AI: add enable_thinking reasoning toggle plumbed to llama.cpp

New optional SamplingOverride forwarded to llama-server as
chat_template_kwargs.enable_thinking (gates Qwen3-style reasoning
blocks). None leaves the template default; other backends ignore it.
Wired through the agentic-insight and chat-turn request bodies/handlers.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

Cargo.lock

@@ -2051,7 +2051,7 @@ dependencies = [
 
 [[package]]
 name = "image-api"
-version = "1.5.0"
+version = "1.4.0"
 dependencies = [
  "actix",
  "actix-cors",

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "image-api"
-version = "1.5.0"
+version = "1.4.0"
 authors = ["Cameron Cordes <cameronc.dev@gmail.com>"]
 edition = "2024"
 

migrations/2026-08-21-000000_persona_chat/down.sql

@@ -1 +0,0 @@
-DROP TABLE IF EXISTS persona_chat_conversations;

migrations/2026-08-21-000000_persona_chat/up.sql

@@ -1,22 +0,0 @@
--- Open chat with persona — file-anchored insight chat is keyed by
--- (library_id, file_path). The open chat is keyed by (user_id, persona_id)
--- with a single rolling transcript per pair. No tree branching in v1
--- (matches the locked-in "single rolling conversation per persona" scope);
--- a flat JSON blob is enough and avoids forcing a tree shape onto a
--- surface that's intentionally linear.
---
--- `messages_json` is the same `Vec<ChatMessage>` shape the file-anchored
--- chat persists, so the SSE replay / UI rendering on the mobile client
--- can reuse the same parser without a second schema.
-
-CREATE TABLE persona_chat_conversations (
-    user_id INTEGER NOT NULL,
-    persona_id TEXT NOT NULL,
-    messages_json TEXT NOT NULL DEFAULT '[]',
-    turn_count INTEGER NOT NULL DEFAULT 0,
-    updated_at BIGINT NOT NULL,
-    PRIMARY KEY (user_id, persona_id)
-);
-
-CREATE INDEX idx_persona_chat_updated
-    ON persona_chat_conversations (user_id, updated_at DESC);

migrations/2026-08-25-000000_persona_chat_multi/down.sql

@@ -1,32 +0,0 @@
--- Collapse back to one conversation per (user, persona). Where a persona has
--- several, the most recently updated one wins and the rest are dropped —
--- the v1 schema has nowhere to put them.
-
-CREATE TABLE persona_chat_conversations_old (
-    user_id INTEGER NOT NULL,
-    persona_id TEXT NOT NULL,
-    messages_json TEXT NOT NULL DEFAULT '[]',
-    turn_count INTEGER NOT NULL DEFAULT 0,
-    updated_at BIGINT NOT NULL,
-    PRIMARY KEY (user_id, persona_id)
-);
-
-INSERT INTO persona_chat_conversations_old (
-    user_id, persona_id, messages_json, turn_count, updated_at
-)
-SELECT user_id, persona_id, messages_json, turn_count, updated_at
-FROM persona_chat_conversations c
-WHERE c.updated_at = (
-    SELECT MAX(c2.updated_at)
-    FROM persona_chat_conversations c2
-    WHERE c2.user_id = c.user_id AND c2.persona_id = c.persona_id
-)
-GROUP BY user_id, persona_id;
-
-DROP INDEX IF EXISTS idx_persona_chat_updated;
-DROP INDEX IF EXISTS idx_persona_chat_persona;
-DROP TABLE persona_chat_conversations;
-ALTER TABLE persona_chat_conversations_old RENAME TO persona_chat_conversations;
-
-CREATE INDEX idx_persona_chat_updated
-    ON persona_chat_conversations (user_id, updated_at DESC);

migrations/2026-08-25-000000_persona_chat_multi/up.sql

@@ -1,50 +0,0 @@
--- Multiple conversations per persona.
---
--- v1 keyed a transcript on (user_id, persona_id), so a persona had exactly
--- one rolling conversation and there was no way to start a fresh topic
--- without discarding the old one. The key is now an opaque `conversation_id`,
--- with (user_id, persona_id) demoted to an index.
---
--- `title` is a short generated summary of the opening exchange, used as the
--- conversation's name in the list. Empty until the first turn completes; the
--- client falls back to the persona name while it is blank.
---
--- SQLite cannot redefine a primary key in place, so this is the standard
--- create-copy-drop-rename dance. Existing transcripts carry over with a
--- generated id and an empty title.
-
-CREATE TABLE persona_chat_conversations_new (
-    conversation_id TEXT NOT NULL PRIMARY KEY,
-    user_id INTEGER NOT NULL,
-    persona_id TEXT NOT NULL,
-    title TEXT NOT NULL DEFAULT '',
-    messages_json TEXT NOT NULL DEFAULT '[]',
-    turn_count INTEGER NOT NULL DEFAULT 0,
-    created_at BIGINT NOT NULL,
-    updated_at BIGINT NOT NULL
-);
-
-INSERT INTO persona_chat_conversations_new (
-    conversation_id, user_id, persona_id, title,
-    messages_json, turn_count, created_at, updated_at
-)
-SELECT
-    lower(hex(randomblob(16))),
-    user_id,
-    persona_id,
-    '',
-    messages_json,
-    turn_count,
-    updated_at,
-    updated_at
-FROM persona_chat_conversations;
-
-DROP INDEX IF EXISTS idx_persona_chat_updated;
-DROP TABLE persona_chat_conversations;
-ALTER TABLE persona_chat_conversations_new RENAME TO persona_chat_conversations;
-
-CREATE INDEX idx_persona_chat_updated
-    ON persona_chat_conversations (user_id, updated_at DESC);
-
-CREATE INDEX idx_persona_chat_persona
-    ON persona_chat_conversations (user_id, persona_id);

src/ai/handlers.rs

@@ -1310,11 +1310,6 @@ pub struct ChatHistoryQuery {
     pub path: String,
     #[serde(default)]
     pub library: Option<String>,
-    /// When set, load the branch anchored at this leaf ID instead of the
-    /// active branch. Allows the client to preview alternate conversation
-    /// paths without making them active.
-    #[serde(default)]
-    pub branch_id: Option<u64>,
 }
 
 #[derive(Debug, Serialize)]
@@ -1323,21 +1318,6 @@ pub struct ChatHistoryHttpResponse {
     pub turn_count: usize,
     pub model_version: String,
     pub backend: String,
-    pub active_leaf_id: u64,
-    /// The branch leaf ID being viewed. Equals `active_leaf_id` when viewing
-    /// the active branch, or the `branch_id` query param for alternates.
-    pub viewing_branch_id: u64,
-    #[serde(skip_serializing_if = "Vec::is_empty")]
-    pub fork_info: Vec<Option<ChatForkInfo>>,
-}
-
-#[derive(Debug, Clone, Serialize)]
-pub struct ChatForkInfo {
-    pub position: usize,
-    pub total: usize,
-    /// The divergence tree node. Pass as `node_id` to the branches endpoint
-    /// for a list scoped to this exact fork.
-    pub node_id: u64,
 }
 
 #[derive(Debug, Serialize)]
@@ -1434,50 +1414,31 @@ pub async fn chat_history_handler(
         .flatten()
         .unwrap_or_else(|| app_state.primary_library());
 
-    match app_state
-        .insight_chat
-        .load_history(library.id, &query.path, query.branch_id)
-    {
-        Ok(view) => {
-            let fork_info: Vec<Option<ChatForkInfo>> = view
-                .fork_info
+    match app_state.insight_chat.load_history(library.id, &query.path) {
+        Ok(view) => HttpResponse::Ok().json(ChatHistoryHttpResponse {
+            messages: view
+                .messages
                 .into_iter()
-                .map(|f| {
-                    f.map(|fi| ChatForkInfo {
-                        position: fi.position,
-                        total: fi.total,
-                        node_id: fi.node_id,
-                    })
+                .map(|m| RenderedHistoryMessage {
+                    role: m.role,
+                    content: m.content,
+                    is_initial: m.is_initial,
+                    tools: m
+                        .tools
+                        .into_iter()
+                        .map(|t| HistoryToolInvocation {
+                            name: t.name,
+                            arguments: t.arguments,
+                            result: t.result,
+                            result_truncated: t.result_truncated,
+                        })
+                        .collect(),
                 })
-                .collect();
-            HttpResponse::Ok().json(ChatHistoryHttpResponse {
-                messages: view
-                    .messages
-                    .into_iter()
-                    .map(|m| RenderedHistoryMessage {
-                        role: m.role,
-                        content: m.content,
-                        is_initial: m.is_initial,
-                        tools: m
-                            .tools
-                            .into_iter()
-                            .map(|t| HistoryToolInvocation {
-                                name: t.name,
-                                arguments: t.arguments,
-                                result: t.result,
-                                result_truncated: t.result_truncated,
-                            })
-                            .collect(),
-                    })
-                    .collect(),
-                turn_count: view.turn_count,
-                model_version: view.model_version,
-                backend: view.backend,
-                active_leaf_id: view.active_leaf_id,
-                viewing_branch_id: view.viewing_branch_id,
-                fork_info,
-            })
-        }
+                .collect(),
+            turn_count: view.turn_count,
+            model_version: view.model_version,
+            backend: view.backend,
+        }),
         Err(e) => {
             let msg = format!("{}", e);
             if msg.contains("no insight found") {
@@ -1491,118 +1452,6 @@ pub async fn chat_history_handler(
     }
 }
 
-/// Query for GET /insights/chat/branches.
-#[derive(Debug, Deserialize)]
-pub struct ChatBranchesQuery {
-    pub path: String,
-    #[serde(default)]
-    pub library: Option<String>,
-    /// When set (from `ForkInfo.node_id`), scope the list to the sibling
-    /// branches diverging at this tree node instead of every leaf.
-    #[serde(default)]
-    pub node_id: Option<u64>,
-    /// The leaf the client is currently viewing. Scoped options in the
-    /// same subtree anchor to it so the client can identify its own branch.
-    /// Defaults to the active leaf when absent.
-    #[serde(default)]
-    pub viewing_branch_id: Option<u64>,
-}
-
-/// GET /insights/chat/branches — return the branch list for a photo's
-/// conversation tree. Without `node_id`, each entry is a leaf node;
-/// with `node_id`, entries are the position-ranked sibling branches at
-/// that divergence point. Use an entry's `id` with `branch_id` on the
-/// history endpoint (or the switch-branch endpoint) to load that path.
-#[get("/insights/chat/branches")]
-pub async fn chat_branches_handler(
-    _claims: Claims,
-    query: web::Query<ChatBranchesQuery>,
-    app_state: web::Data<AppState>,
-) -> impl Responder {
-    let library = libraries::resolve_library_param_state(&app_state, query.library.as_deref())
-        .ok()
-        .flatten()
-        .unwrap_or_else(|| app_state.primary_library());
-
-    let (branches, active_leaf_id) = match app_state.insight_chat.get_branches(
-        library.id,
-        &query.path,
-        query.node_id,
-        query.viewing_branch_id,
-    ) {
-        Ok(result) => result,
-        Err(e) => {
-            let msg = format!("{}", e);
-            if msg.contains("no insight found") {
-                return HttpResponse::NotFound().json(serde_json::json!({ "error": msg }));
-            } else if msg.contains("no chat history") {
-                return HttpResponse::Conflict().json(serde_json::json!({ "error": msg }));
-            } else if msg.contains("not found in tree") {
-                return HttpResponse::BadRequest().json(serde_json::json!({ "error": msg }));
-            } else {
-                return HttpResponse::InternalServerError()
-                    .json(serde_json::json!({ "error": msg }));
-            }
-        }
-    };
-
-    HttpResponse::Ok().json(serde_json::json!({
-        "branches": branches,
-        "active_leaf_id": active_leaf_id,
-    }))
-}
-
-/// POST /insights/chat/switch-branch — switch the active branch to a
-/// different conversation path. The new branch becomes the active
-/// conversation; the previous active branch becomes a regular fork.
-#[post("/insights/chat/switch-branch")]
-pub async fn chat_switch_branch_handler(
-    _claims: Claims,
-    request: web::Json<ChatSwitchBranchRequest>,
-    app_state: web::Data<AppState>,
-) -> impl Responder {
-    let library =
-        match libraries::resolve_library_param_state(&app_state, request.library.as_deref()) {
-            Ok(Some(lib)) => lib,
-            Ok(None) => app_state.primary_library(),
-            Err(e) => {
-                return HttpResponse::BadRequest().json(serde_json::json!({
-                    "error": format!("invalid library: {}", e)
-                }));
-            }
-        };
-
-    match app_state
-        .insight_chat
-        .switch_branch(library.id, &request.file_path, request.branch_id)
-        .await
-    {
-        Ok(()) => HttpResponse::Ok().json(serde_json::json!({ "success": true })),
-        Err(e) => {
-            let msg = format!("{}", e);
-            log::error!("Switch branch failed: {}", msg);
-            if msg.contains("no insight found") {
-                HttpResponse::NotFound().json(serde_json::json!({ "error": msg }))
-            } else if msg.contains("no chat history") {
-                HttpResponse::Conflict().json(serde_json::json!({ "error": msg }))
-            } else if msg.contains("not found") || msg.contains("not a leaf") {
-                HttpResponse::BadRequest().json(serde_json::json!({ "error": msg }))
-            } else {
-                HttpResponse::InternalServerError().json(serde_json::json!({ "error": msg }))
-            }
-        }
-    }
-}
-
-#[derive(Debug, Deserialize)]
-pub struct ChatSwitchBranchRequest {
-    pub file_path: String,
-    #[serde(default)]
-    pub library: Option<String>,
-    /// The leaf node ID to switch to. Must be a valid leaf in the tree.
-    pub branch_id: u64,
-}
-
 /// POST /insights/chat/stream — streaming variant of /insights/chat.
 /// Returns `text/event-stream` with one event per chat stream event.
 #[post("/insights/chat/stream")]
@@ -1819,19 +1668,6 @@ pub async fn turn_replay_handler(
     path: web::Path<String>,
     query: web::Query<ReplayQuery>,
     app_state: web::Data<AppState>,
-) -> HttpResponse {
-    turn_replay_impl(http_request, path, query, app_state).await
-}
-
-/// Core of the SSE replay, kept attribute-free so the persona-chat routes
-/// can reuse it under `/persona_chat/turn/{turn_id}`. The registry is keyed
-/// on `turn_id` only, and `render_turn_info_frame` already scopes the
-/// identity (persona turns carry `persona_id` and no `file_path`).
-pub(crate) async fn turn_replay_impl(
-    http_request: HttpRequest,
-    path: web::Path<String>,
-    query: web::Query<ReplayQuery>,
-    app_state: web::Data<AppState>,
 ) -> HttpResponse {
     use crate::ai::turn_registry::ReplayOutcome;
 
@@ -1962,31 +1798,15 @@ pub(crate) async fn turn_replay_impl(
 }
 
 fn render_turn_info_frame(info: &crate::ai::turn_registry::TurnInfo) -> String {
-    // Persona-scoped turns leak no `file_path` to the client — the open
-    // chat is keyed on `(persona_id)` only and a stray path echo from a
-    // quoted SMS would be a privacy regression. Insight-scoped turns
-    // include the path as before.
-    let mut payload = serde_json::Map::new();
-    payload.insert("turn_id".into(), serde_json::json!(info.turn_id));
-    payload.insert("library_id".into(), serde_json::json!(info.library_id));
-    payload.insert("scope".into(), serde_json::json!(info.scope));
-    payload.insert("status".into(), serde_json::json!(info.status.as_str()));
-    payload.insert(
-        "total_events_pushed".into(),
-        serde_json::json!(info.total_events_pushed),
-    );
-    payload.insert(
-        "buffered_count".into(),
-        serde_json::json!(info.buffered_count),
-    );
-    if info.scope == "insight" {
-        payload.insert("file_path".into(), serde_json::json!(info.file_path));
-    }
-    if let Some(ref pid) = info.persona_id {
-        payload.insert("persona_id".into(), serde_json::json!(pid));
-    }
-    let data = serde_json::to_string(&serde_json::Value::Object(payload))
-        .unwrap_or_else(|_| "{}".to_string());
+    let payload = serde_json::json!({
+        "turn_id": info.turn_id,
+        "file_path": info.file_path,
+        "library_id": info.library_id,
+        "status": info.status.as_str(),
+        "total_events_pushed": info.total_events_pushed,
+        "buffered_count": info.buffered_count,
+    });
+    let data = serde_json::to_string(&payload).unwrap_or_else(|_| "{}".to_string());
     format!("event: turn_info\ndata: {}\n\n", data)
 }
 
@@ -1996,16 +1816,6 @@ pub async fn cancel_turn_handler(
     http_request: HttpRequest,
     path: web::Path<String>,
     app_state: web::Data<AppState>,
-) -> impl Responder {
-    cancel_turn_impl(http_request, path, app_state).await
-}
-
-/// Core of the turn-cancel, attribute-free so the persona-chat DELETE route
-/// can reuse it under `/persona_chat/turn/{turn_id}`.
-pub(crate) async fn cancel_turn_impl(
-    http_request: HttpRequest,
-    path: web::Path<String>,
-    app_state: web::Data<AppState>,
 ) -> impl Responder {
     let turn_id = path.into_inner();
 
@@ -2056,13 +1866,6 @@ pub(crate) async fn cancel_turn_impl(
     }))
 }
 
-// Persona-chat handlers (`/persona_chat/*`) live in `ai::persona_chat` and
-// are registered in main.rs. The persona SSE replay and cancel routes are
-// thin wrappers over `turn_replay_handler` + `cancel_turn_handler`: both
-// are keyed on `turn_id`, not file_path, and `TurnInfo::scope` ("persona")
-// lets the shared `turn_info` frame carry the right identity (persona_id,
-// no file_path).
-
 #[cfg(test)]
 mod turn_replay_tests {
     use super::{cancel_turn_handler, render_indexed_frame, turn_replay_handler};

src/ai/insight_generator.rs

@@ -26,7 +26,7 @@ use crate::libraries::Library;
 use crate::memories::extract_date_from_filename;
 use crate::otel::global_tracer;
 use crate::tags::TagDao;
-use crate::utils::{earliest_fs_time, normalize_path, retry_with_backoff};
+use crate::utils::{earliest_fs_time, normalize_path};
 
 /// Max location records rendered by `tool_get_location_history`. The DAO
 /// query is range-bounded, not limited, so the tool caps the rendered list
@@ -1539,13 +1539,10 @@ impl InsightGenerator {
             eval_count: None,
         };
 
-        let dao = self.insight_dao.clone();
-        let result = retry_with_backoff("store_insight", 3, || {
-            let cx = opentelemetry::Context::new();
-            let mut d = dao.lock().expect("Unable to lock InsightDao");
-            d.store_insight(&cx, insight.clone())
-        })
-        .map_err(|e| anyhow::anyhow!("Failed to store insight: {:?}", e));
+        let mut dao = self.insight_dao.lock().expect("Unable to lock InsightDao");
+        let result = dao
+            .store_insight(&insight_cx, insight)
+            .map_err(|e| anyhow::anyhow!("Failed to store insight: {:?}", e));
 
         match &result {
             Ok(_) => {
@@ -4468,13 +4465,11 @@ Return ONLY the summary, nothing else."#,
             eval_count: last_eval_count,
         };
 
-        let dao = self.insight_dao.clone();
-        let stored = retry_with_backoff("store_insight", 3, || {
-            let cx = opentelemetry::Context::new();
-            let mut d = dao.lock().expect("Unable to lock InsightDao");
-            d.store_insight(&cx, insight.clone())
-        })
-        .map_err(|e| anyhow::anyhow!("Failed to store agentic insight: {:?}", e));
+        let stored = {
+            let mut dao = self.insight_dao.lock().expect("Unable to lock InsightDao");
+            dao.store_insight(&insight_cx, insight)
+                .map_err(|e| anyhow::anyhow!("Failed to store agentic insight: {:?}", e))
+        };
 
         match &stored {
             Ok(_) => {

src/ai/llm_client.rs

@@ -171,292 +171,6 @@ pub struct ModelCapabilities {
     pub has_tool_calling: bool,
 }
 
-/// A single node in the conversation tree. Wraps ChatMessage with structural
-/// metadata so the tree can be persisted as JSON in `training_messages`.
-#[derive(Serialize, Deserialize, Clone, Debug)]
-pub struct StoredChatNode {
-    pub id: u64,
-    #[serde(skip_serializing_if = "Option::is_none")]
-    pub parent_id: Option<u64>,
-    pub message: ChatMessage,
-}
-
-/// Top-level container for tree-based chat history. Stored as JSON in the
-/// `training_messages` column. Backward-compatible reader detects whether
-/// the stored value is a flat array (old format) or this object (new format).
-#[derive(Serialize, Deserialize, Clone, Debug)]
-pub struct ChatHistoryStore {
-    pub nodes: Vec<StoredChatNode>,
-    pub active_leaf_id: u64,
-}
-
-impl ChatHistoryStore {
-    /// Create an empty store. Test-only: production stores are always built
-    /// from `from_flat_array` or deserialized from `training_messages`.
-    #[cfg(test)]
-    pub fn empty() -> Self {
-        Self {
-            nodes: Vec::new(),
-            active_leaf_id: 0,
-        }
-    }
-
-    /// Convert a flat `Vec<ChatMessage>` (old format) into a tree.
-    /// Each message becomes a node chained by parent_id.
-    pub fn from_flat_array(messages: Vec<ChatMessage>) -> Self {
-        let mut nodes = Vec::with_capacity(messages.len());
-        let mut prev_id: Option<u64> = None;
-        for (i, msg) in messages.into_iter().enumerate() {
-            let id = (i + 1) as u64;
-            nodes.push(StoredChatNode {
-                id,
-                parent_id: prev_id,
-                message: msg,
-            });
-            prev_id = Some(id);
-        }
-        let active_leaf_id = prev_id.unwrap_or(0);
-        Self {
-            nodes,
-            active_leaf_id,
-        }
-    }
-
-    /// Allocate the next node ID (max existing + 1, or 1 if empty).
-    pub fn next_id(&self) -> u64 {
-        self.nodes.iter().map(|n| n.id).max().unwrap_or(0) + 1
-    }
-
-    /// Append a new node as a child of the given parent. Returns the new node's id.
-    pub fn append_node(&mut self, parent_id: Option<u64>, message: ChatMessage) -> u64 {
-        let id = self.next_id();
-        self.nodes.push(StoredChatNode {
-            id,
-            parent_id,
-            message,
-        });
-        id
-    }
-
-    /// Traverse from root to the given leaf_id, returning the path in order.
-    /// Returns None if the leaf_id doesn't exist.
-    pub fn path_to_leaf(&self, leaf_id: u64) -> Option<Vec<&StoredChatNode>> {
-        // Build a map from id to node for fast lookup.
-        let node_map: std::collections::HashMap<u64, usize> = self
-            .nodes
-            .iter()
-            .enumerate()
-            .map(|(i, n)| (n.id, i))
-            .collect();
-        let &_leaf_idx = node_map.get(&leaf_id)?;
-
-        let mut path_indices = Vec::new();
-        let mut current_id = leaf_id;
-        while let Some(&idx) = node_map.get(&current_id) {
-            path_indices.push(idx);
-            match self.nodes[idx].parent_id {
-                Some(pid) => current_id = pid,
-                None => break,
-            }
-        }
-        path_indices.reverse();
-        Some(path_indices.into_iter().map(|i| &self.nodes[i]).collect())
-    }
-
-    /// All direct children of a given node, ordered by id (chronological).
-    pub fn children_of(&self, node_id: u64) -> Vec<&StoredChatNode> {
-        self.nodes
-            .iter()
-            .filter(|n| n.parent_id == Some(node_id))
-            .collect()
-    }
-
-    /// For a given node, return (position, total) where:
-    /// - total = number of children of the parent of this node (i.e., siblings + self)
-    /// - position = 1-based rank of this node among siblings, ordered by id
-    ///
-    /// Returns None if the node has no parent (root) or the parent has 0 or 1 child.
-    pub fn fork_at_node(&self, node_id: u64) -> Option<(usize, usize)> {
-        let node = self.nodes.iter().find(|n| n.id == node_id)?;
-        let parent_id = node.parent_id?;
-        let children = self.children_of(parent_id);
-        let total = children.len();
-        if total <= 1 {
-            return None;
-        }
-        let position = children.iter().position(|c| c.id == node_id)? + 1;
-        Some((position, total))
-    }
-
-    /// All leaf node IDs within the subtree rooted at `node_id` (the node
-    /// itself when it has no children). Empty when the node doesn't exist.
-    pub fn descendant_leaves(&self, node_id: u64) -> Vec<u64> {
-        if !self.nodes.iter().any(|n| n.id == node_id) {
-            return Vec::new();
-        }
-        let mut result = Vec::new();
-        let mut stack = vec![node_id];
-        while let Some(id) = stack.pop() {
-            let children = self.children_of(id);
-            if children.is_empty() {
-                result.push(id);
-            } else {
-                stack.extend(children.iter().map(|c| c.id));
-            }
-        }
-        result
-    }
-
-    /// Branch options at a divergence point: one entry per direct child of
-    /// `fork_node_id`, ordered by id (chronological — matches the stable
-    /// position rank shown in "X/Y" fork chips).
-    ///
-    /// Each entry's `id` is the leaf to switch to for that branch: the
-    /// `preferred_leaf` when it lives in that child's subtree (so the
-    /// caller's current view maps onto its own option and can be filtered
-    /// out client-side), otherwise the most recent (max-id) descendant leaf.
-    ///
-    /// Snippets preview the first visible message that *differs* between
-    /// the branches, not blindly the first child: Rewind & Regenerate
-    /// resends the identical question, so the first child is often shared
-    /// across siblings and would make every option read the same.
-    pub fn branch_options_at(&self, fork_node_id: u64, preferred_leaf: u64) -> Vec<BranchLeafInfo> {
-        struct Candidate {
-            rep: u64,
-            message_count: usize,
-            /// User/assistant contents after the divergence, tool
-            /// scaffolding (tool results, empty dispatch turns) skipped.
-            visible: Vec<String>,
-        }
-        let candidates: Vec<Candidate> = self
-            .children_of(fork_node_id)
-            .iter()
-            .map(|child| {
-                let leaves = self.descendant_leaves(child.id);
-                let rep = if leaves.contains(&preferred_leaf) {
-                    preferred_leaf
-                } else {
-                    leaves.iter().copied().max().unwrap_or(child.id)
-                };
-                let path = self.path_to_leaf(rep).unwrap_or_default();
-                let message_count = path.len();
-                let visible: Vec<String> = path
-                    .iter()
-                    .skip_while(|n| n.id != child.id)
-                    .filter(|n| {
-                        (n.message.role == "user" || n.message.role == "assistant")
-                            && !n.message.content.trim().is_empty()
-                    })
-                    .map(|n| n.message.content.clone())
-                    .collect();
-                Candidate {
-                    rep,
-                    message_count,
-                    visible,
-                }
-            })
-            .collect();
-
-        // Walk forward past positions where every branch has the same
-        // content; stop at the first divergence (or when any branch runs
-        // out — a length difference is itself distinguishing).
-        let mut snippet_idx = 0usize;
-        loop {
-            let contents: Vec<Option<&String>> = candidates
-                .iter()
-                .map(|c| c.visible.get(snippet_idx))
-                .collect();
-            let all_present_and_equal =
-                contents.iter().all(|c| c.is_some()) && contents.windows(2).all(|w| w[0] == w[1]);
-            if !all_present_and_equal {
-                break;
-            }
-            snippet_idx += 1;
-        }
-
-        candidates
-            .into_iter()
-            .enumerate()
-            .map(|(i, c)| {
-                // A branch exhausted before the divergence index falls back
-                // to its own last message so it still shows *something*.
-                let snippet = c
-                    .visible
-                    .get(snippet_idx)
-                    .or_else(|| c.visible.last())
-                    .map(|s| s.chars().take(100).collect::<String>())
-                    .unwrap_or_default();
-                BranchLeafInfo {
-                    id: c.rep,
-                    snippet,
-                    message_count: c.message_count,
-                    position: Some(i + 1),
-                }
-            })
-            .collect()
-    }
-
-    /// All leaf node IDs (nodes with no children).
-    pub fn leaves(&self) -> Vec<u64> {
-        let child_ids: std::collections::HashSet<u64> =
-            self.nodes.iter().filter_map(|n| n.parent_id).collect();
-        self.nodes
-            .iter()
-            .filter(|n| !child_ids.contains(&n.id))
-            .map(|n| n.id)
-            .collect()
-    }
-
-    /// The parent of the given node, if any. Used to re-anchor a rewind
-    /// that discards every rendered message onto the seed node above them.
-    pub fn parent_of(&self, node_id: u64) -> Option<&StoredChatNode> {
-        let node = self.nodes.iter().find(|n| n.id == node_id)?;
-        let parent_id = node.parent_id?;
-        self.nodes.iter().find(|n| n.id == parent_id)
-    }
-
-    /// Return all leaf nodes with metadata: id, snippet (first 100 chars of
-    /// the first user-visible message after the divergence point), and
-    /// message_count (number of nodes in the path from root to leaf).
-    pub fn leaves_with_info(&self) -> Vec<BranchLeafInfo> {
-        let leaf_ids = self.leaves();
-        leaf_ids
-            .into_iter()
-            .map(|leaf_id| {
-                let path = self.path_to_leaf(leaf_id).unwrap_or_default();
-                let message_count = path.len();
-                // Snippet: first 100 chars of the first user/assistant message
-                // (skip system messages).
-                let snippet = path
-                    .iter()
-                    .find(|n| n.message.role != "system")
-                    .map(|n| n.message.content.chars().take(100).collect::<String>())
-                    .unwrap_or_default();
-                BranchLeafInfo {
-                    id: leaf_id,
-                    snippet,
-                    message_count,
-                    position: None,
-                }
-            })
-            .collect()
-    }
-}
-
-/// Metadata about a branch leaf, returned by the branches endpoint.
-#[derive(Serialize, Clone, Debug)]
-pub struct BranchLeafInfo {
-    pub id: u64,
-    #[serde(skip_serializing_if = "String::is_empty")]
-    pub snippet: String,
-    pub message_count: usize,
-    /// 1-based sibling rank at the divergence point. Only present for
-    /// scoped queries (`branch_options_at`); tree-wide leaf listings have
-    /// no single divergence to rank against.
-    #[serde(skip_serializing_if = "Option::is_none")]
-    pub position: Option<usize>,
-}
-
 /// Strip a leading `<think>…</think>` reasoning block from model output.
 ///
 /// Thinking models sometimes emit chain-of-thought inside think tags before
@@ -507,284 +221,4 @@ mod tests {
         let raw = "<think>thinking forever";
         assert_eq!(strip_think_blocks(raw), raw);
     }
-
-    // ── ChatHistoryStore tree traversal tests ──────────────────────────
-
-    fn mk_msg(role: &str, content: &str) -> ChatMessage {
-        match role {
-            "system" => ChatMessage::system(content),
-            "user" => ChatMessage::user(content),
-            _ => ChatMessage {
-                role: role.to_string(),
-                content: content.to_string(),
-                tool_calls: None,
-                images: None,
-            },
-        }
-    }
-
-    #[test]
-    fn store_from_flat_array_chains_parent_ids() {
-        let msgs = vec![
-            mk_msg("system", "sys"),
-            mk_msg("user", "hello"),
-            mk_msg("assistant", "hi there"),
-        ];
-        let store = ChatHistoryStore::from_flat_array(msgs);
-        assert_eq!(store.nodes.len(), 3);
-        assert_eq!(store.nodes[0].id, 1);
-        assert_eq!(store.nodes[0].parent_id, None);
-        assert_eq!(store.nodes[1].parent_id, Some(1));
-        assert_eq!(store.nodes[2].parent_id, Some(2));
-        assert_eq!(store.active_leaf_id, 3);
-    }
-
-    #[test]
-    fn store_path_to_leaf_returns_ordered_path() {
-        let mut store = ChatHistoryStore::empty();
-        let n1 = store.append_node(None, mk_msg("user", "u1"));
-        let n2 = store.append_node(Some(n1), mk_msg("assistant", "a1"));
-        let n3 = store.append_node(Some(n2), mk_msg("user", "u2"));
-        let path = store.path_to_leaf(n3).unwrap();
-        assert_eq!(path.len(), 3);
-        assert_eq!(path[0].id, n1);
-        assert_eq!(path[1].id, n2);
-        assert_eq!(path[2].id, n3);
-    }
-
-    #[test]
-    fn store_path_to_leaf_returns_none_for_missing_id() {
-        let store = ChatHistoryStore::empty();
-        assert!(store.path_to_leaf(999).is_none());
-    }
-
-    #[test]
-    fn store_children_of_returns_direct_children() {
-        let mut store = ChatHistoryStore::empty();
-        let root = store.append_node(None, mk_msg("user", "root"));
-        let c1 = store.append_node(Some(root), mk_msg("assistant", "c1"));
-        let c2 = store.append_node(Some(root), mk_msg("user", "c2"));
-        let _d1 = store.append_node(Some(c1), mk_msg("assistant", "d1"));
-        let children = store.children_of(root);
-        assert_eq!(children.len(), 2);
-        assert_eq!(children[0].id, c1);
-        assert_eq!(children[1].id, c2);
-    }
-
-    #[test]
-    fn store_fork_at_node_detects_fork() {
-        let mut store = ChatHistoryStore::empty();
-        let root = store.append_node(None, mk_msg("user", "root"));
-        let c1 = store.append_node(Some(root), mk_msg("assistant", "c1"));
-        let c2 = store.append_node(Some(root), mk_msg("user", "c2"));
-        let c3 = store.append_node(Some(root), mk_msg("user", "c3"));
-        // c1 is position 1 of 3 siblings
-        assert_eq!(store.fork_at_node(c1), Some((1, 3)));
-        assert_eq!(store.fork_at_node(c2), Some((2, 3)));
-        assert_eq!(store.fork_at_node(c3), Some((3, 3)));
-        // root has no parent, so no fork
-        assert!(store.fork_at_node(root).is_none());
-    }
-
-    #[test]
-    fn store_fork_at_node_returns_none_for_single_child() {
-        let mut store = ChatHistoryStore::empty();
-        let root = store.append_node(None, mk_msg("user", "root"));
-        let _only = store.append_node(Some(root), mk_msg("assistant", "only"));
-        assert!(store.fork_at_node(2).is_none());
-    }
-
-    #[test]
-    fn store_leaves_returns_leaf_ids() {
-        let mut store = ChatHistoryStore::empty();
-        let root = store.append_node(None, mk_msg("user", "root"));
-        let c1 = store.append_node(Some(root), mk_msg("assistant", "c1"));
-        let c2 = store.append_node(Some(root), mk_msg("user", "c2"));
-        let _d1 = store.append_node(Some(c1), mk_msg("assistant", "d1"));
-        let leaves = store.leaves();
-        assert_eq!(leaves.len(), 2);
-        assert!(leaves.contains(&c2));
-        assert!(leaves.contains(&_d1));
-        assert!(!leaves.contains(&root));
-        assert!(!leaves.contains(&c1));
-    }
-
-    #[test]
-    fn store_parent_of_returns_parent() {
-        let mut store = ChatHistoryStore::empty();
-        let root = store.append_node(None, mk_msg("user", "root"));
-        let child = store.append_node(Some(root), mk_msg("assistant", "child"));
-        let parent = store.parent_of(child).unwrap();
-        assert_eq!(parent.id, root);
-        assert!(store.parent_of(root).is_none());
-    }
-
-    #[test]
-    fn store_next_id_increments() {
-        let mut store = ChatHistoryStore::empty();
-        assert_eq!(store.next_id(), 1);
-        store.append_node(None, mk_msg("user", "a"));
-        assert_eq!(store.next_id(), 2);
-        store.append_node(Some(1), mk_msg("assistant", "b"));
-        assert_eq!(store.next_id(), 3);
-    }
-
-    #[test]
-    fn store_descendant_leaves_walks_subtree() {
-        let mut store = ChatHistoryStore::empty();
-        let root = store.append_node(None, mk_msg("user", "root"));
-        let c1 = store.append_node(Some(root), mk_msg("assistant", "c1"));
-        let c2 = store.append_node(Some(root), mk_msg("user", "c2"));
-        let d1 = store.append_node(Some(c1), mk_msg("assistant", "d1"));
-        let d2 = store.append_node(Some(c1), mk_msg("assistant", "d2"));
-        // c1's subtree has two leaves; c2 is its own leaf; root sees all.
-        let mut c1_leaves = store.descendant_leaves(c1);
-        c1_leaves.sort_unstable();
-        assert_eq!(c1_leaves, vec![d1, d2]);
-        assert_eq!(store.descendant_leaves(c2), vec![c2]);
-        let mut all = store.descendant_leaves(root);
-        all.sort_unstable();
-        assert_eq!(all, vec![c2, d1, d2]);
-        assert!(store.descendant_leaves(999).is_empty());
-    }
-
-    #[test]
-    fn branch_options_ranks_children_and_prefers_viewing_leaf() {
-        let mut store = ChatHistoryStore::empty();
-        let root = store.append_node(None, mk_msg("user", "root"));
-        // First branch: two leaves (older + newer).
-        let c1 = store.append_node(Some(root), mk_msg("assistant", "reply one"));
-        let old_leaf = store.append_node(Some(c1), mk_msg("user", "old follow-up"));
-        let new_leaf = store.append_node(Some(c1), mk_msg("user", "new follow-up"));
-        // Second branch: single leaf.
-        let c2 = store.append_node(Some(root), mk_msg("assistant", "reply two"));
-
-        // Viewing the *older* leaf of branch 1: its option must anchor to
-        // that exact leaf, not the max-id one, so the client's
-        // "hide the branch I'm on" filter matches.
-        let opts = store.branch_options_at(root, old_leaf);
-        assert_eq!(opts.len(), 2);
-        assert_eq!(opts[0].id, old_leaf);
-        assert_eq!(opts[0].position, Some(1));
-        assert_eq!(opts[0].snippet, "reply one");
-        assert_eq!(opts[1].id, c2);
-        assert_eq!(opts[1].position, Some(2));
-
-        // Viewing branch 2: branch 1 falls back to its most recent leaf.
-        let opts = store.branch_options_at(root, c2);
-        assert_eq!(opts[0].id, new_leaf);
-        assert_eq!(opts[1].id, c2);
-    }
-
-    #[test]
-    fn branch_options_snippet_skips_shared_resent_question() {
-        // Rewind & Regenerate resends the identical question, so both
-        // branches' first child matches — the preview must advance to the
-        // first *differing* message (the assistant replies) or every
-        // option would read the same.
-        let mut store = ChatHistoryStore::empty();
-        let root = store.append_node(None, mk_msg("assistant", "prior reply"));
-        let q1 = store.append_node(Some(root), mk_msg("user", "same question"));
-        let _a1 = store.append_node(Some(q1), mk_msg("assistant", "first answer"));
-        let q2 = store.append_node(Some(root), mk_msg("user", "same question"));
-        let a2 = store.append_node(Some(q2), mk_msg("assistant", "second answer"));
-
-        let opts = store.branch_options_at(root, a2);
-        assert_eq!(opts.len(), 2);
-        assert_eq!(opts[0].snippet, "first answer");
-        assert_eq!(opts[1].snippet, "second answer");
-    }
-
-    #[test]
-    fn branch_options_snippet_falls_back_when_branch_exhausted() {
-        // One branch is just the shared question (no reply yet); the other
-        // continues past it. The shorter branch falls back to its own last
-        // message rather than showing an empty preview.
-        let mut store = ChatHistoryStore::empty();
-        let root = store.append_node(None, mk_msg("assistant", "prior reply"));
-        let _q1 = store.append_node(Some(root), mk_msg("user", "same question"));
-        let q2 = store.append_node(Some(root), mk_msg("user", "same question"));
-        let a2 = store.append_node(Some(q2), mk_msg("assistant", "second answer"));
-
-        let opts = store.branch_options_at(root, a2);
-        assert_eq!(opts[0].snippet, "same question");
-        assert_eq!(opts[1].snippet, "second answer");
-    }
-
-    #[test]
-    fn branch_options_snippet_skips_tool_scaffolding() {
-        let mut store = ChatHistoryStore::empty();
-        let root = store.append_node(None, mk_msg("user", "root"));
-        // Branch whose diverging child is an empty tool-dispatch turn.
-        let dispatch = store.append_node(
-            Some(root),
-            ChatMessage {
-                role: "assistant".to_string(),
-                content: String::new(),
-                tool_calls: None,
-                images: None,
-            },
-        );
-        let tool = store.append_node(Some(dispatch), mk_msg("tool", "raw tool output"));
-        let reply = store.append_node(Some(tool), mk_msg("assistant", "final answer"));
-        let _other = store.append_node(Some(root), mk_msg("assistant", "direct reply"));
-
-        let opts = store.branch_options_at(root, reply);
-        assert_eq!(opts[0].snippet, "final answer");
-        assert_eq!(opts[1].snippet, "direct reply");
-    }
-
-    #[test]
-    fn store_append_node_sets_parent() {
-        let mut store = ChatHistoryStore::empty();
-        let parent = store.append_node(None, mk_msg("user", "parent"));
-        let child = store.append_node(Some(parent), mk_msg("assistant", "child"));
-        assert_eq!(store.nodes.len(), 2);
-        assert_eq!(store.nodes[1].parent_id, Some(parent));
-        assert_eq!(store.nodes[1].id, child);
-    }
-
-    #[test]
-    fn branch_options_handles_nested_forks() {
-        // Regression: verify sub-fork display when there are multiple forks.
-        // Structure:
-        //   root (user "q1")
-        //   ├── a1 (assistant "answer 1")
-        //   └── a2 (assistant "answer 2")
-        //       ├── q2 (user "q2")
-        //       │   ├── r1 (assistant "reply 1")
-        //       │   └── r2 (assistant "reply 2")  <- active leaf
-        //
-        // branch_options_at(root) should return [a1, a2] (2 branches)
-        // branch_options_at(q2) should return [r1, r2] (2 sub-branches)
-        let mut store = ChatHistoryStore::empty();
-        let root = store.append_node(None, mk_msg("user", "q1"));
-        let a1 = store.append_node(Some(root), mk_msg("assistant", "answer 1"));
-        let a2 = store.append_node(Some(root), mk_msg("assistant", "answer 2"));
-        let q2 = store.append_node(Some(a2), mk_msg("user", "q2"));
-        let r1 = store.append_node(Some(q2), mk_msg("assistant", "reply 1"));
-        let r2 = store.append_node(Some(q2), mk_msg("assistant", "reply 2"));
-        store.active_leaf_id = r2;
-
-        // Root fork: should show [a1, a2's subtree]
-        let opts = store.branch_options_at(root, r2);
-        assert_eq!(opts.len(), 2);
-        assert_eq!(opts[0].id, a1);
-        assert_eq!(opts[0].position, Some(1));
-        assert_eq!(opts[0].snippet, "answer 1");
-        // a2's subtree includes r2, so the rep should be r2 (preferred leaf)
-        assert_eq!(opts[1].id, r2);
-        assert_eq!(opts[1].position, Some(2));
-        assert_eq!(opts[1].snippet, "answer 2");
-
-        // Sub-fork at q2: should show [r1, r2]
-        let opts = store.branch_options_at(q2, r2);
-        assert_eq!(opts.len(), 2);
-        assert_eq!(opts[0].id, r1);
-        assert_eq!(opts[0].position, Some(1));
-        assert_eq!(opts[0].snippet, "reply 1");
-        assert_eq!(opts[1].id, r2);
-        assert_eq!(opts[1].position, Some(2));
-        assert_eq!(opts[1].snippet, "reply 2");
-    }
 }

src/ai/mod.rs

@@ -13,7 +13,6 @@ pub mod local_llm;
 pub mod nl_query;
 pub mod ollama;
 pub mod openrouter;
-pub mod persona_chat;
 pub mod pronunciation;
 pub mod sms_client;
 pub mod tts;
@@ -26,12 +25,12 @@ pub use daily_summary_job::{
     generate_daily_summaries, strip_summary_boilerplate,
 };
 pub use handlers::{
-    cancel_generation_handler, cancel_turn_handler, chat_branches_handler, chat_history_handler,
-    chat_rewind_handler, chat_stream_handler, chat_switch_branch_handler, chat_turn_handler,
-    delete_insight_handler, export_training_data_handler, generate_agentic_insight_handler,
-    generate_insight_handler, generation_status_handler, get_all_insights_handler,
-    get_available_models_handler, get_insight_handler, get_insight_history_handler,
-    get_openrouter_models_handler, rate_insight_handler, turn_async_handler, turn_replay_handler,
+    cancel_generation_handler, cancel_turn_handler, chat_history_handler, chat_rewind_handler,
+    chat_stream_handler, chat_turn_handler, delete_insight_handler, export_training_data_handler,
+    generate_agentic_insight_handler, generate_insight_handler, generation_status_handler,
+    get_all_insights_handler, get_available_models_handler, get_insight_handler,
+    get_insight_history_handler, get_openrouter_models_handler, rate_insight_handler,
+    turn_async_handler, turn_replay_handler,
 };
 pub use insight_generator::InsightGenerator;
 pub use llamacpp::LlamaCppClient;
@@ -39,12 +38,6 @@ pub use llamacpp::LlamaCppClient;
 pub use llm_client::{
     ChatMessage, LlmClient, ModelCapabilities, Tool, ToolCall, ToolCallFunction, ToolFunction,
 };
-pub use persona_chat::{
-    persona_chat_branches_handler, persona_chat_conversations_handler,
-    persona_chat_create_conversation_handler, persona_chat_delete_conversation_handler,
-    persona_chat_history_handler, persona_chat_rewind_handler, persona_chat_switch_branch_handler,
-    persona_chat_turn_handler, persona_turn_cancel_handler, persona_turn_replay_handler,
-};
 // LocalLlm is constructed by binaries (reembed_embeddings, importers), not the server
 #[allow(unused_imports)]
 pub use local_llm::LocalLlm;

src/ai/turn_registry.rs

@@ -52,10 +52,6 @@ pub struct TurnInfo {
     pub turn_id: String,
     pub file_path: String,
     pub library_id: i32,
-    /// Persona id for persona-scoped turns. None for insight-scoped turns.
-    pub persona_id: Option<String>,
-    /// `"insight"` (file-anchored chat) | `"persona"` (open chat).
-    pub scope: String,
     pub status: TurnStatus,
     pub total_events_pushed: u32,
     pub buffered_count: u32,
@@ -82,20 +78,8 @@ pub enum ReplayOutcome {
 /// replay connections (readers).
 pub struct TurnEntry {
     pub turn_id: String,
-    /// Stable identity used to scope the turn. For insight chat this is
-    /// `(file_path, library_id)`. For persona chat it is `(persona_id)`.
-    /// The generic fields stay populated regardless of scope so the SSE
-    /// `turn_info` frame can always surface a useful label — a future
-    /// "list active turns" debugging surface won't need to discriminate.
     pub file_path: String,
     pub library_id: i32,
-    /// Persona id for persona-scoped turns (`scope == "persona"`). None for
-    /// the existing insight-scoped turns. The field is the cheapest way to
-    /// let the SSE `turn_info` payload round-trip without a parallel map.
-    pub persona_id: Option<String>,
-    /// `"insight"` (default — file-anchored chat) | `"persona"` (open chat).
-    /// Stable string so the SSE `turn_info` payload can carry it as-is.
-    pub scope: String,
     /// Shared event buffer — multiple SSE connections can read independently.
     /// Each connection tracks its own `skip_before` offset.
     events: Mutex<Vec<ChatStreamEvent>>,
@@ -116,35 +100,10 @@ pub struct TurnEntry {
 
 impl TurnEntry {
     pub fn new(turn_id: String, file_path: String, library_id: i32) -> Self {
-        Self::with_scope(turn_id, file_path, library_id, None, "insight")
-    }
-
-    /// Persona-scoped constructor. `file_path` is unused by the persona
-    /// chat loop but kept on the struct so the SSE `turn_info` frame has
-    /// the same shape across scopes.
-    pub fn new_persona(turn_id: String, persona_id: String, library_id: i32) -> Self {
-        Self::with_scope(
-            turn_id,
-            String::new(),
-            library_id,
-            Some(persona_id),
-            "persona",
-        )
-    }
-
-    fn with_scope(
-        turn_id: String,
-        file_path: String,
-        library_id: i32,
-        persona_id: Option<String>,
-        scope: &str,
-    ) -> Self {
         Self {
             turn_id,
             file_path,
             library_id,
-            persona_id,
-            scope: scope.to_string(),
             events: Mutex::new(Vec::new()),
             total_events_pushed: AtomicU32::new(0),
             base_index: AtomicU32::new(0),
@@ -211,8 +170,6 @@ impl TurnEntry {
             turn_id: self.turn_id.clone(),
             file_path: self.file_path.clone(),
             library_id: self.library_id,
-            persona_id: self.persona_id.clone(),
-            scope: self.scope.clone(),
             status: self.status.load(Ordering::Relaxed).into(),
             total_events_pushed: total,
             buffered_count: buffered,
@@ -788,67 +745,4 @@ mod tests {
         let from_base = events_of(entry.replay_from(5).await);
         assert_eq!(from_base.len(), MAX_BUFFERED_EVENTS);
     }
-
-    // ── Persona-scope regression guards ───────────────────────────
-    //
-    // The open-chat surface relies on `scope == "persona"` and the
-    // optional `persona_id` surviving a round-trip through `info()`.
-    // A future refactor that hard-codes "insight" or drops the field
-    // would silently break the SSE `turn_info` payload — guard it.
-
-    #[tokio::test]
-    async fn turn_entry_supports_persona_scope_label() {
-        let entry = Arc::new(TurnEntry::new_persona(
-            "tp1".to_string(),
-            "journal".to_string(),
-            1,
-        ));
-        assert_eq!(entry.scope, "persona");
-        assert_eq!(entry.persona_id.as_deref(), Some("journal"));
-        let info = entry.info().await;
-        assert_eq!(info.scope, "persona");
-        assert_eq!(info.persona_id.as_deref(), Some("journal"));
-        // File-anchored fields stay blank — the persona chat has no file.
-        assert!(info.file_path.is_empty());
-    }
-
-    #[tokio::test]
-    async fn turn_info_payload_for_persona_does_not_leak_file_path() {
-        // Defensive: even if some future code path accidentally populates
-        // file_path on a persona entry, the SSE `turn_info` payload
-        // should not surface it to clients. Today file_path is always
-        // empty for persona entries, so we assert that contract.
-        let entry = Arc::new(TurnEntry::new_persona(
-            "tp2".to_string(),
-            "default".to_string(),
-            1,
-        ));
-        let info = entry.info().await;
-        let json = serde_json::to_string(&serde_json::json!({
-            "turn_id": info.turn_id,
-            "persona_id": info.persona_id,
-            "scope": info.scope,
-            "status": info.status.as_str(),
-            "total_events_pushed": info.total_events_pushed,
-            "buffered_count": info.buffered_count,
-        }))
-        .unwrap();
-        // Privacy: a persona chat reply that quotes an SMS must not echo
-        // the file path that message originally surfaced from.
-        assert!(
-            !json.contains("file_path"),
-            "persona turn_info payload should not leak file_path: {json}"
-        );
-    }
-
-    #[tokio::test]
-    async fn insight_scope_defaults_when_using_legacy_constructor() {
-        // The legacy `TurnEntry::new` constructor (used by the existing
-        // file-anchored insight chat) must still report scope="insight"
-        // so callers can discriminate without a parallel map.
-        let entry = Arc::new(TurnEntry::new("t-insight".into(), "/p.jpg".into(), 1));
-        let info = entry.info().await;
-        assert_eq!(info.scope, "insight");
-        assert!(info.persona_id.is_none());
-    }
 }

src/database/models.rs

@@ -192,7 +192,7 @@ pub struct ImageExif {
     pub clip_model_version: Option<String>,
 }
 
-#[derive(Insertable, Clone)]
+#[derive(Insertable)]
 #[diesel(table_name = photo_insights)]
 pub struct InsertPhotoInsight {
     pub library_id: i32,

src/database/persona_dao.rs

@@ -10,51 +10,6 @@ use crate::database::schema;
 use crate::database::{DbError, DbErrorKind, connect};
 use crate::otel::trace_db_call;
 
-/// One row of the persona-chat transcript. Lives in
-/// `persona_chat_conversations` (one row per `(user_id, persona_id)`).
-///
-/// One persona conversation.
-///
-/// `turn_count` is the cumulative number of assistant turns on the active
-/// branch — what the list screen shows as the length of a conversation. The
-/// transcript itself is the `messages_json` blob, holding a serialized
-/// `ChatHistoryStore` tree. `title` is a generated summary of the opening
-/// exchange, empty until the first turn completes.
-#[derive(Clone, Debug)]
-pub struct PersonaChatRow {
-    pub conversation_id: String,
-    pub persona_id: String,
-    pub title: String,
-    pub messages_json: String,
-    pub turn_count: i32,
-    pub created_at: i64,
-    pub updated_at: i64,
-}
-
-/// Column tuple → `PersonaChatRow`. Shared by the single-row and list
-/// queries so their `select(...)` orders can never drift apart.
-fn persona_chat_row(
-    (conversation_id, persona_id, title, messages_json, turn_count, created_at, updated_at): (
-        String,
-        String,
-        String,
-        String,
-        i32,
-        i64,
-        i64,
-    ),
-) -> PersonaChatRow {
-    PersonaChatRow {
-        conversation_id,
-        persona_id,
-        title,
-        messages_json,
-        turn_count,
-        created_at,
-        updated_at,
-    }
-}
-
 /// Patch shape for update_persona. None = leave field alone. Built-ins are
 /// allowed to flip `include_all_memories` but should reject name/prompt
 /// edits at the handler layer (built-in copy lives in the migration).
@@ -125,76 +80,6 @@ pub trait PersonaDao: Sync + Send {
         user_id: i32,
         personas: &[ImportPersona],
     ) -> Result<usize, DbError>;
-
-    // ── Persona-chat (open chat with persona) persistence ───────────
-    //
-    // Keyed on an opaque `conversation_id` so one persona can hold several
-    // separate conversations. `user_id` is checked on every read and write:
-    // the id is a bearer token for a transcript, so ownership can never be
-    // assumed from the id alone.
-
-    /// Fetch one conversation. None when it doesn't exist or belongs to
-    /// another user — the two are deliberately indistinguishable to callers.
-    fn get_persona_chat(
-        &mut self,
-        cx: &opentelemetry::Context,
-        user_id: i32,
-        conversation_id: &str,
-    ) -> Result<Option<PersonaChatRow>, DbError>;
-
-    /// Every conversation this user has going, newest first. Backs the chat
-    /// list screen, so it returns the transcript blob too — the snippet is
-    /// derived from the tree rather than denormalized into its own column.
-    fn list_persona_chats(
-        &mut self,
-        cx: &opentelemetry::Context,
-        user_id: i32,
-    ) -> Result<Vec<PersonaChatRow>, DbError>;
-
-    /// Start a new conversation with a persona, returning its id. Several
-    /// conversations with the same persona are expected, so this never
-    /// reuses an existing row.
-    fn create_persona_chat(
-        &mut self,
-        cx: &opentelemetry::Context,
-        user_id: i32,
-        persona_id: &str,
-        created_at: i64,
-    ) -> Result<String, DbError>;
-
-    /// Replace a conversation's transcript. Called once per completed turn
-    /// with the full new `messages_json`; `turn_count` is the cumulative
-    /// assistant-turn count on the resulting active branch. Returns the
-    /// number of rows written — 0 means the conversation is gone or is not
-    /// this user's.
-    fn update_persona_chat(
-        &mut self,
-        cx: &opentelemetry::Context,
-        user_id: i32,
-        conversation_id: &str,
-        messages_json: &str,
-        turn_count: i32,
-        updated_at: i64,
-    ) -> Result<usize, DbError>;
-
-    /// Name a conversation. Written once, after the first turn produces
-    /// enough of an exchange to summarize.
-    fn set_persona_chat_title(
-        &mut self,
-        cx: &opentelemetry::Context,
-        user_id: i32,
-        conversation_id: &str,
-        title: &str,
-    ) -> Result<(), DbError>;
-
-    /// Delete a conversation outright. Backs both the list screen's delete
-    /// affordance and the reset endpoint.
-    fn delete_persona_chat(
-        &mut self,
-        cx: &opentelemetry::Context,
-        user_id: i32,
-        conversation_id: &str,
-    ) -> Result<(), DbError>;
 }
 
 pub struct SqlitePersonaDao {
@@ -366,28 +251,12 @@ impl PersonaDao for SqlitePersonaDao {
         pid: &str,
     ) -> Result<bool, DbError> {
         trace_db_call(cx, "delete", "delete_persona", |_span| {
+            use schema::personas::dsl::*;
             let mut conn = self.connection.lock().expect("PersonaDao lock");
-            // One transaction so a persona and its conversations can't get
-            // out of step. There is no FK between the two tables, and the
-            // chat list skips conversations whose persona is gone — so
-            // without this the transcripts would linger invisibly forever.
-            conn.deref_mut().transaction::<_, anyhow::Error, _>(|tx| {
-                {
-                    use schema::persona_chat_conversations::dsl::*;
-                    diesel::delete(
-                        persona_chat_conversations
-                            .filter(user_id.eq(uid))
-                            .filter(persona_id.eq(pid)),
-                    )
-                    .execute(tx)
-                    .map_err(|e| anyhow::anyhow!("Delete error: {}", e))?;
-                }
-                use schema::personas::dsl::*;
-                let n = diesel::delete(personas.filter(user_id.eq(uid)).filter(persona_id.eq(pid)))
-                    .execute(tx)
-                    .map_err(|e| anyhow::anyhow!("Delete error: {}", e))?;
-                Ok(n > 0)
-            })
+            let n = diesel::delete(personas.filter(user_id.eq(uid)).filter(persona_id.eq(pid)))
+                .execute(conn.deref_mut())
+                .map_err(|e| anyhow::anyhow!("Delete error: {}", e))?;
+            Ok(n > 0)
         })
         .map_err(|e| DbError::log(DbErrorKind::QueryError, e))
     }
@@ -427,168 +296,8 @@ impl PersonaDao for SqlitePersonaDao {
         })
         .map_err(|e| DbError::log(DbErrorKind::InsertError, e))
     }
-
-    fn get_persona_chat(
-        &mut self,
-        cx: &opentelemetry::Context,
-        uid: i32,
-        cid: &str,
-    ) -> Result<Option<PersonaChatRow>, DbError> {
-        trace_db_call(cx, "query", "get_persona_chat", |_span| {
-            use schema::persona_chat_conversations::dsl::*;
-            let mut conn = self.connection.lock().expect("PersonaDao lock");
-            persona_chat_conversations
-                .filter(conversation_id.eq(cid))
-                // Scoped by user as well as id: the id is a bearer token for
-                // someone's transcript, so it never grants access on its own.
-                .filter(user_id.eq(uid))
-                .select((
-                    conversation_id,
-                    persona_id,
-                    title,
-                    messages_json,
-                    turn_count,
-                    created_at,
-                    updated_at,
-                ))
-                .first::<(String, String, String, String, i32, i64, i64)>(conn.deref_mut())
-                .optional()
-                .map(|opt| opt.map(persona_chat_row))
-                .map_err(|e| anyhow::anyhow!("Query error: {}", e))
-        })
-        .map_err(|e| DbError::log(DbErrorKind::QueryError, e))
-    }
-
-    fn list_persona_chats(
-        &mut self,
-        cx: &opentelemetry::Context,
-        uid: i32,
-    ) -> Result<Vec<PersonaChatRow>, DbError> {
-        trace_db_call(cx, "query", "list_persona_chats", |_span| {
-            use schema::persona_chat_conversations::dsl::*;
-            let mut conn = self.connection.lock().expect("PersonaDao lock");
-            persona_chat_conversations
-                .filter(user_id.eq(uid))
-                .order(updated_at.desc())
-                .select((
-                    conversation_id,
-                    persona_id,
-                    title,
-                    messages_json,
-                    turn_count,
-                    created_at,
-                    updated_at,
-                ))
-                .load::<(String, String, String, String, i32, i64, i64)>(conn.deref_mut())
-                .map(|rows| rows.into_iter().map(persona_chat_row).collect())
-                .map_err(|e| anyhow::anyhow!("Query error: {}", e))
-        })
-        .map_err(|e| DbError::log(DbErrorKind::QueryError, e))
-    }
-
-    fn create_persona_chat(
-        &mut self,
-        cx: &opentelemetry::Context,
-        uid: i32,
-        pid: &str,
-        created: i64,
-    ) -> Result<String, DbError> {
-        trace_db_call(cx, "insert", "create_persona_chat", |_span| {
-            use schema::persona_chat_conversations::dsl::*;
-            let mut conn = self.connection.lock().expect("PersonaDao lock");
-            let new_id = uuid::Uuid::new_v4().to_string();
-            diesel::insert_into(persona_chat_conversations)
-                .values((
-                    conversation_id.eq(&new_id),
-                    user_id.eq(uid),
-                    persona_id.eq(pid),
-                    title.eq(""),
-                    // The empty tree; the first turn seeds it with the
-                    // persona's system prompt.
-                    messages_json.eq("[]"),
-                    turn_count.eq(0),
-                    created_at.eq(created),
-                    updated_at.eq(created),
-                ))
-                .execute(conn.deref_mut())
-                .map_err(|e| anyhow::anyhow!("Insert error: {}", e))?;
-            Ok(new_id)
-        })
-        .map_err(|e| DbError::log(DbErrorKind::InsertError, e))
-    }
-
-    fn update_persona_chat(
-        &mut self,
-        cx: &opentelemetry::Context,
-        uid: i32,
-        cid: &str,
-        json: &str,
-        count: i32,
-        updated: i64,
-    ) -> Result<usize, DbError> {
-        trace_db_call(cx, "update", "update_persona_chat", |_span| {
-            use schema::persona_chat_conversations::dsl::*;
-            let mut conn = self.connection.lock().expect("PersonaDao lock");
-            diesel::update(
-                persona_chat_conversations
-                    .filter(conversation_id.eq(cid))
-                    .filter(user_id.eq(uid)),
-            )
-            .set((
-                messages_json.eq(json),
-                turn_count.eq(count),
-                updated_at.eq(updated),
-            ))
-            .execute(conn.deref_mut())
-            .map_err(|e| anyhow::anyhow!("Update error: {}", e))
-        })
-        .map_err(|e| DbError::log(DbErrorKind::QueryError, e))
-    }
-
-    fn set_persona_chat_title(
-        &mut self,
-        cx: &opentelemetry::Context,
-        uid: i32,
-        cid: &str,
-        new_title: &str,
-    ) -> Result<(), DbError> {
-        trace_db_call(cx, "update", "set_persona_chat_title", |_span| {
-            use schema::persona_chat_conversations::dsl::*;
-            let mut conn = self.connection.lock().expect("PersonaDao lock");
-            diesel::update(
-                persona_chat_conversations
-                    .filter(conversation_id.eq(cid))
-                    .filter(user_id.eq(uid)),
-            )
-            .set(title.eq(new_title))
-            .execute(conn.deref_mut())
-            .map_err(|e| anyhow::anyhow!("Update error: {}", e))?;
-            Ok(())
-        })
-        .map_err(|e| DbError::log(DbErrorKind::QueryError, e))
-    }
-
-    fn delete_persona_chat(
-        &mut self,
-        cx: &opentelemetry::Context,
-        uid: i32,
-        cid: &str,
-    ) -> Result<(), DbError> {
-        trace_db_call(cx, "delete", "delete_persona_chat", |_span| {
-            use schema::persona_chat_conversations::dsl::*;
-            let mut conn = self.connection.lock().expect("PersonaDao lock");
-            diesel::delete(
-                persona_chat_conversations
-                    .filter(conversation_id.eq(cid))
-                    .filter(user_id.eq(uid)),
-            )
-            .execute(conn.deref_mut())
-            .map_err(|e| anyhow::anyhow!("Delete error: {}", e))?;
-            Ok(())
-        })
-        .map_err(|e| DbError::log(DbErrorKind::QueryError, e))
-    }
 }
+
 #[cfg(test)]
 mod tests {
     use super::*;
@@ -735,201 +444,4 @@ mod tests {
             .unwrap();
         assert!(updated.include_all_memories);
     }
-
-    // ── Persona-chat DAO tests ─────────────────────────────────────
-
-    /// Second user, for the isolation tests.
-    fn second_user(dao: &SqlitePersonaDao, username: &str) -> i32 {
-        use crate::database::schema::users::dsl as u;
-        let conn = dao.connection.clone();
-        diesel::insert_into(u::users)
-            .values((u::username.eq(username), u::password.eq("x")))
-            .execute(conn.lock().unwrap().deref_mut())
-            .unwrap();
-        u::users
-            .filter(u::username.eq(username))
-            .select(u::id)
-            .first(conn.lock().unwrap().deref_mut())
-            .unwrap()
-    }
-
-    #[test]
-    fn persona_chat_get_returns_none_for_an_unknown_conversation() {
-        let cx = opentelemetry::Context::new();
-        let (mut dao, uid) = dao_with_user("p1");
-        assert!(
-            dao.get_persona_chat(&cx, uid, "no-such-id")
-                .unwrap()
-                .is_none()
-        );
-    }
-
-    #[test]
-    fn persona_chat_create_then_get_round_trip() {
-        let cx = opentelemetry::Context::new();
-        let (mut dao, uid) = dao_with_user("p2");
-        let cid = dao.create_persona_chat(&cx, uid, "journal", 100).unwrap();
-
-        let row = dao.get_persona_chat(&cx, uid, &cid).unwrap().unwrap();
-        assert_eq!(row.conversation_id, cid);
-        assert_eq!(row.persona_id, "journal");
-        assert_eq!(row.title, "", "unnamed until the first turn completes");
-        assert_eq!(row.turn_count, 0);
-        assert_eq!(row.created_at, 100);
-        assert_eq!(row.updated_at, 100);
-    }
-
-    #[test]
-    fn persona_chat_create_makes_a_separate_row_each_time() {
-        // The whole point of the conversation_id key: one persona, several
-        // independent threads.
-        let cx = opentelemetry::Context::new();
-        let (mut dao, uid) = dao_with_user("p3");
-        let first = dao.create_persona_chat(&cx, uid, "journal", 100).unwrap();
-        let second = dao.create_persona_chat(&cx, uid, "journal", 200).unwrap();
-
-        assert_ne!(first, second);
-        assert_eq!(dao.list_persona_chats(&cx, uid).unwrap().len(), 2);
-    }
-
-    #[test]
-    fn persona_chat_update_replaces_the_transcript() {
-        let cx = opentelemetry::Context::new();
-        let (mut dao, uid) = dao_with_user("p4");
-        let cid = dao.create_persona_chat(&cx, uid, "journal", 100).unwrap();
-
-        let rows = dao
-            .update_persona_chat(&cx, uid, &cid, "first", 1, 150)
-            .unwrap();
-        assert_eq!(rows, 1);
-        let rows = dao
-            .update_persona_chat(&cx, uid, &cid, "second", 2, 200)
-            .unwrap();
-        assert_eq!(rows, 1);
-
-        let row = dao.get_persona_chat(&cx, uid, &cid).unwrap().unwrap();
-        assert_eq!(row.messages_json, "second");
-        assert_eq!(row.turn_count, 2);
-        assert_eq!(row.updated_at, 200);
-        assert_eq!(row.created_at, 100, "creation time is not disturbed");
-    }
-
-    #[test]
-    fn persona_chat_update_reports_zero_rows_for_an_unknown_conversation() {
-        // The turn loop treats 0 as "the conversation went away mid-turn"
-        // rather than silently succeeding.
-        let cx = opentelemetry::Context::new();
-        let (mut dao, uid) = dao_with_user("p5");
-        let rows = dao
-            .update_persona_chat(&cx, uid, "no-such-id", "x", 1, 1)
-            .unwrap();
-        assert_eq!(rows, 0);
-    }
-
-    #[test]
-    fn persona_chat_set_title_names_the_conversation() {
-        let cx = opentelemetry::Context::new();
-        let (mut dao, uid) = dao_with_user("p6");
-        let cid = dao.create_persona_chat(&cx, uid, "journal", 100).unwrap();
-
-        dao.set_persona_chat_title(&cx, uid, &cid, "June recap")
-            .unwrap();
-        let row = dao.get_persona_chat(&cx, uid, &cid).unwrap().unwrap();
-        assert_eq!(row.title, "June recap");
-    }
-
-    #[test]
-    fn persona_chat_list_orders_newest_first() {
-        let cx = opentelemetry::Context::new();
-        let (mut dao, uid) = dao_with_user("p7");
-        let older = dao.create_persona_chat(&cx, uid, "journal", 100).unwrap();
-        let newer = dao.create_persona_chat(&cx, uid, "coach", 300).unwrap();
-        let middle = dao.create_persona_chat(&cx, uid, "default", 200).unwrap();
-
-        let ids: Vec<String> = dao
-            .list_persona_chats(&cx, uid)
-            .unwrap()
-            .into_iter()
-            .map(|r| r.conversation_id)
-            .collect();
-        assert_eq!(
-            ids,
-            vec![newer, middle, older],
-            "ordered by updated_at descending so the list screen needs no re-sort"
-        );
-    }
-
-    #[test]
-    fn persona_chat_list_is_empty_for_a_user_who_has_never_chatted() {
-        let cx = opentelemetry::Context::new();
-        let (mut dao, uid) = dao_with_user("p8");
-        assert!(dao.list_persona_chats(&cx, uid).unwrap().is_empty());
-    }
-
-    #[test]
-    fn persona_chat_delete_removes_the_conversation() {
-        let cx = opentelemetry::Context::new();
-        let (mut dao, uid) = dao_with_user("p9");
-        let cid = dao.create_persona_chat(&cx, uid, "journal", 100).unwrap();
-
-        dao.delete_persona_chat(&cx, uid, &cid).unwrap();
-        assert!(dao.get_persona_chat(&cx, uid, &cid).unwrap().is_none());
-        assert!(dao.list_persona_chats(&cx, uid).unwrap().is_empty());
-    }
-
-    #[test]
-    fn deleting_a_persona_takes_its_conversations_with_it() {
-        // No FK between the tables, and the chat list hides conversations
-        // whose persona is gone — so an orphan row would be an invisible,
-        // unreachable transcript sitting in the database forever.
-        let cx = opentelemetry::Context::new();
-        let (mut dao, uid) = dao_with_user("p-cascade");
-        dao.create_persona(&cx, uid, "custom-1", "Custom", "prompt", false, false)
-            .unwrap();
-        dao.create_persona_chat(&cx, uid, "custom-1", 100).unwrap();
-        dao.create_persona_chat(&cx, uid, "custom-1", 200).unwrap();
-        let survivor = dao.create_persona_chat(&cx, uid, "other", 300).unwrap();
-
-        assert!(dao.delete_persona(&cx, uid, "custom-1").unwrap());
-
-        let remaining: Vec<String> = dao
-            .list_persona_chats(&cx, uid)
-            .unwrap()
-            .into_iter()
-            .map(|r| r.conversation_id)
-            .collect();
-        assert_eq!(
-            remaining,
-            vec![survivor],
-            "only the untouched persona's conversation survives"
-        );
-    }
-
-    #[test]
-    fn persona_chat_reads_and_writes_are_scoped_to_the_owner() {
-        // A conversation_id is a bearer token for someone's transcript, so
-        // holding one must not grant another user access to it.
-        let cx = opentelemetry::Context::new();
-        let (mut dao, uid1) = dao_with_user("owner");
-        let uid2 = second_user(&dao, "intruder");
-        let cid = dao.create_persona_chat(&cx, uid1, "journal", 100).unwrap();
-        dao.update_persona_chat(&cx, uid1, &cid, "private", 1, 100)
-            .unwrap();
-
-        assert!(
-            dao.get_persona_chat(&cx, uid2, &cid).unwrap().is_none(),
-            "another user cannot read it"
-        );
-        assert_eq!(
-            dao.update_persona_chat(&cx, uid2, &cid, "tampered", 9, 999)
-                .unwrap(),
-            0,
-            "another user cannot write it"
-        );
-        dao.delete_persona_chat(&cx, uid2, &cid).unwrap();
-
-        let row = dao.get_persona_chat(&cx, uid1, &cid).unwrap().unwrap();
-        assert_eq!(row.messages_json, "private", "and cannot delete it");
-        assert!(dao.list_persona_chats(&cx, uid2).unwrap().is_empty());
-    }
 }

src/database/schema.rs

@@ -171,19 +171,6 @@ diesel::table! {
     }
 }
 
-diesel::table! {
-    persona_chat_conversations (conversation_id) {
-        conversation_id -> Text,
-        user_id -> Integer,
-        persona_id -> Text,
-        title -> Text,
-        messages_json -> Text,
-        turn_count -> Integer,
-        created_at -> BigInt,
-        updated_at -> BigInt,
-    }
-}
-
 diesel::table! {
     personas (id) {
         id -> Integer,
@@ -358,7 +345,6 @@ diesel::allow_tables_to_appear_in_same_query!(
     insight_generation_jobs,
     libraries,
     location_history,
-    persona_chat_conversations,
     personas,
     persons,
     photo_insights,

src/main.rs

@@ -377,21 +377,9 @@ fn main() -> std::io::Result<()> {
                 .service(ai::chat_stream_handler)
                 .service(ai::chat_history_handler)
                 .service(ai::chat_rewind_handler)
-                .service(ai::chat_branches_handler)
-                .service(ai::chat_switch_branch_handler)
                 .service(ai::turn_async_handler)
                 .service(ai::turn_replay_handler)
                 .service(ai::cancel_turn_handler)
-                .service(ai::persona_chat_history_handler)
-                .service(ai::persona_chat_turn_handler)
-                .service(ai::persona_chat_create_conversation_handler)
-                .service(ai::persona_chat_delete_conversation_handler)
-                .service(ai::persona_turn_replay_handler)
-                .service(ai::persona_turn_cancel_handler)
-                .service(ai::persona_chat_rewind_handler)
-                .service(ai::persona_chat_switch_branch_handler)
-                .service(ai::persona_chat_branches_handler)
-                .service(ai::persona_chat_conversations_handler)
                 .service(ai::rate_insight_handler)
                 .service(ai::export_training_data_handler)
                 .service(ai::tts_speech_handler)

src/state.rs

@@ -4,7 +4,6 @@ use crate::ai::face_client::FaceClient;
 use crate::ai::insight_chat::{ChatLockMap, InsightChatService};
 use crate::ai::llamacpp::LlamaCppClient;
 use crate::ai::openrouter::OpenRouterClient;
-use crate::ai::persona_chat::PersonaChatSession;
 use crate::ai::turn_registry::TurnRegistry;
 use crate::ai::{InsightGenerator, OllamaClient, SmsApiClient};
 use crate::database::{
@@ -85,9 +84,6 @@ pub struct AppState {
     pub insight_generator: InsightGenerator,
     /// Chat continuation service. Hold an Arc so handlers can clone cheaply.
     pub insight_chat: Arc<InsightChatService>,
-    /// Open chat with persona service. Same shape as insight_chat but
-    /// anchored on (user_id, persona_id) instead of (library_id, file_path).
-    pub persona_chat_session: Arc<PersonaChatSession>,
     pub turn_registry: Arc<TurnRegistry>,
     pub face_client: FaceClient,
     pub clip_client: ClipClient,
@@ -137,7 +133,6 @@ impl AppState {
         sms_client: SmsApiClient,
         insight_generator: InsightGenerator,
         insight_chat: Arc<InsightChatService>,
-        persona_chat_session: Arc<PersonaChatSession>,
         turn_registry: Arc<TurnRegistry>,
         preview_dao: Arc<Mutex<Box<dyn PreviewDao>>>,
         face_client: FaceClient,
@@ -199,7 +194,6 @@ impl AppState {
             sms_client,
             insight_generator,
             insight_chat,
-            persona_chat_session,
             turn_registry,
             face_client,
             clip_client,
@@ -322,7 +316,7 @@ impl Default for AppState {
             tag_dao.clone(),
             face_dao.clone(),
             knowledge_dao,
-            persona_dao.clone(),
+            persona_dao,
             libraries_vec.clone(),
         );
 
@@ -336,13 +330,6 @@ impl Default for AppState {
             chat_locks,
         ));
 
-        // Open chat with persona: reuses the generator + persona DAO.
-        let persona_chat_session = Arc::new(PersonaChatSession::new(
-            Arc::new(insight_generator.clone()),
-            persona_dao.clone(),
-            Arc::new(tokio::sync::Mutex::new(std::collections::HashMap::new())),
-        ));
-
         // Turn registry for reconnectable chat turns. 5-minute timeout for
         // stale turns (background cleaner drops entries older than this).
         let timeout_secs: u64 = env::var("INSIGHT_CHAT_TURN_TIMEOUT_SECS")
@@ -373,7 +360,6 @@ impl Default for AppState {
             sms_client,
             insight_generator,
             insight_chat,
-            persona_chat_session,
             turn_registry,
             preview_dao,
             face_client,
@@ -542,7 +528,7 @@ impl AppState {
             tag_dao.clone(),
             face_dao.clone(),
             knowledge_dao,
-            persona_dao.clone(),
+            persona_dao,
             vec![test_lib],
         );
 
@@ -554,13 +540,6 @@ impl AppState {
             chat_locks,
         ));
 
-        // Open chat with persona (test).
-        let persona_chat_session = Arc::new(PersonaChatSession::new(
-            Arc::new(insight_generator.clone()),
-            persona_dao.clone(),
-            Arc::new(tokio::sync::Mutex::new(std::collections::HashMap::new())),
-        ));
-
         // Turn registry for test state.
         let turn_registry = Arc::new(TurnRegistry::new(300));
 
@@ -592,7 +571,6 @@ impl AppState {
             sms_client,
             insight_generator,
             insight_chat,
-            persona_chat_session,
             turn_registry,
             preview_dao,
             FaceClient::new(None), // disabled in test

src/thumbnails.rs

@@ -98,7 +98,7 @@ pub fn generate_image_thumbnail(src: &Path, thumb_path: &Path) -> std::io::Resul
     }
 
     if file_types::needs_ffmpeg_thumbnail(src) {
-        return generate_image_thumbnail_ffmpeg(src, thumb_path, orientation);
+        return generate_image_thumbnail_ffmpeg(src, thumb_path);
     }
 
     let img = image::open(src).map_err(|e| {
@@ -140,7 +140,7 @@ pub fn generate_large_preview(src: &Path, dest: &Path) -> std::io::Result<()> {
     }
 
     if file_types::needs_ffmpeg_thumbnail(src) {
-        return generate_large_preview_ffmpeg(src, dest, orientation);
+        return generate_large_preview_ffmpeg(src, dest);
     }
 
     let img = image::open(src).map_err(|e| {
@@ -175,30 +175,14 @@ fn encode_large_jpeg(img: image::DynamicImage, dest: &Path) -> std::io::Result<(
 /// ffmpeg path for HEIC/HEIF (image crate can't decode these). Mirrors
 /// [`crate::video::actors::generate_image_thumbnail_ffmpeg`] but scales
 /// to the large-preview cap instead of 200.
-fn generate_large_preview_ffmpeg(src: &Path, dest: &Path, orientation: i32) -> std::io::Result<()> {
-    // Rotation + scale + colorspace. HEIC sources use Display P3; without
-    // colorspace=bt709 the mjpeg encoder treats P3 values as sRGB, producing
-    // warm/oversaturated output. The min(iw,cap) trick caps the long edge
-    // regardless of orientation, mirroring image::thumbnail.
-    let rotation = match orientation {
-        2 => "hflip",
-        3 => "transpose=2",
-        4 => "vflip",
-        5 => "transpose=0,hflip",
-        6 => "transpose=0",
-        7 => "transpose=1,hflip",
-        8 => "transpose=1",
-        _ => "",
-    };
-    let scale_expr = format!(
+fn generate_large_preview_ffmpeg(src: &Path, dest: &Path) -> std::io::Result<()> {
+    // scale=W:-1 with force_original_aspect_ratio=decrease + the min(iw,W)
+    // trick caps the long edge regardless of orientation, mirroring what
+    // image::thumbnail does for the non-ffmpeg branch.
+    let vf = format!(
         "scale='if(gt(iw,ih),min(iw,{cap}),-1)':'if(gt(iw,ih),-1,min(ih,{cap}))'",
         cap = LARGE_PREVIEW_MAX_DIM
     );
-    let vf = if rotation.is_empty() {
-        format!("{},colorspace=bt709", scale_expr)
-    } else {
-        format!("{},{},colorspace=bt709", rotation, scale_expr)
-    };
     let output = Command::new("ffmpeg")
         .arg("-y")
         .arg("-i")
@@ -248,7 +232,7 @@ pub fn generate_xlarge_preview(src: &Path, dest: &Path) -> std::io::Result<()> {
     }
 
     if file_types::needs_ffmpeg_thumbnail(src) {
-        return generate_xlarge_preview_ffmpeg(src, dest, orientation);
+        return generate_xlarge_preview_ffmpeg(src, dest);
     }
 
     let img = image::open(src).map_err(|e| {
@@ -276,30 +260,11 @@ fn encode_xlarge_jpeg(img: image::DynamicImage, dest: &Path) -> std::io::Result<
     Ok(())
 }
 
-fn generate_xlarge_preview_ffmpeg(
-    src: &Path,
-    dest: &Path,
-    orientation: i32,
-) -> std::io::Result<()> {
-    let rotation = match orientation {
-        2 => "hflip",
-        3 => "transpose=2",
-        4 => "vflip",
-        5 => "transpose=0,hflip",
-        6 => "transpose=0",
-        7 => "transpose=1,hflip",
-        8 => "transpose=1",
-        _ => "",
-    };
-    let scale_expr = format!(
+fn generate_xlarge_preview_ffmpeg(src: &Path, dest: &Path) -> std::io::Result<()> {
+    let vf = format!(
         "scale='if(gt(iw,ih),min(iw,{cap}),-1)':'if(gt(iw,ih),-1,min(ih,{cap}))'",
         cap = XLARGE_PREVIEW_MAX_DIM
     );
-    let vf = if rotation.is_empty() {
-        format!("{},colorspace=bt709", scale_expr)
-    } else {
-        format!("{},{},colorspace=bt709", rotation, scale_expr)
-    };
     let output = Command::new("ffmpeg")
         .arg("-y")
         .arg("-i")

src/utils.rs

@@ -1,58 +1,4 @@
-use rand::Rng;
-use std::time::{Duration, SystemTime};
-
-/// Retry a fallible operation with exponential backoff + jitter.
-///
-/// Runs the closure immediately, then retries up to `max_retries` times on
-/// failure. Each retry sleeps for `base_delay * 2^attempt` plus a uniform
-/// jitter of ±25%. Non-final errors are logged at `debug` with the label;
-/// the final error is logged at `error`.
-///
-/// # Examples
-/// ```
-/// use image_api::utils::retry_with_backoff;
-///
-/// let result = retry_with_backoff("my-op", 3, || {
-///     // something that may fail transiently
-///     Ok::<_, anyhow::Error>(42)
-/// });
-/// ```
-pub fn retry_with_backoff<F, T, E>(label: &str, max_retries: u32, mut op: F) -> Result<T, E>
-where
-    F: FnMut() -> Result<T, E>,
-    E: std::fmt::Debug,
-{
-    let mut last_err = match op() {
-        Ok(v) => return Ok(v),
-        Err(e) => e,
-    };
-    for attempt in 1..=max_retries {
-        let base = Duration::from_millis(100).saturating_mul(1_u32.pow(attempt - 1));
-        let jitter_range = base.as_millis() as f64 * 0.25;
-        let jitter = rand::thread_rng().gen_range(-jitter_range..=jitter_range) as u128;
-        let delay = base.as_millis() as i128 + jitter as i128;
-        std::thread::sleep(Duration::from_millis(delay.max(0) as u64));
-        log::debug!(
-            "{}: attempt {}/{} failed ({:?}), retrying in {}ms",
-            label,
-            attempt,
-            max_retries,
-            last_err,
-            delay.max(0)
-        );
-        match op() {
-            Ok(v) => return Ok(v),
-            Err(e) => last_err = e,
-        }
-    }
-    log::error!(
-        "{}: all {} retries exhausted: {:?}",
-        label,
-        max_retries,
-        last_err
-    );
-    Err(last_err)
-}
+use std::time::SystemTime;
 
 /// Normalize a file path to use forward slashes for cross-platform consistency
 /// This ensures paths stored in the database always use `/` regardless of OS

src/video/actors.rs

@@ -90,39 +90,10 @@ pub fn generate_video_thumbnail(path: &Path, destination: &Path) -> std::io::Res
     Ok(())
 }
 
-/// Build the ffmpeg filter chain for image thumbnails: rotation (from EXIF
-/// orientation) + scale + color-space conversion. HEIC sources use Display P3
-/// primaries; without `colorspace=bt709` the mjpeg encoder treats P3 values
-/// as sRGB, producing warm/oversaturated output.
-fn build_image_thumb_filter(orientation: i32, scale_w: u32) -> String {
-    let rotation = match orientation {
-        2 => "hflip",
-        3 => "transpose=2",
-        4 => "vflip",
-        5 => "transpose=0,hflip",
-        6 => "transpose=0",
-        7 => "transpose=1,hflip",
-        8 => "transpose=1",
-        _ => "", // orientation 1 or unknown — no rotation needed
-    };
-    if rotation.is_empty() {
-        format!("scale={}:{{-1}},colorspace=bt709", scale_w)
-    } else {
-        format!("{},scale={}:{{-1}},colorspace=bt709", rotation, scale_w)
-    }
-}
-
-/// Use ffmpeg to extract a thumbnail from formats the `image` crate can't
-/// decode (HEIC/HEIF, RAW: NEF/ARW). `orientation` is the EXIF Orientation
-/// tag value (1..=8) — baked into the pixels so the saved JPEG is
-/// canonically oriented. Writes JPEG bytes to `destination` regardless of
-/// its extension.
-pub fn generate_image_thumbnail_ffmpeg(
-    path: &Path,
-    destination: &Path,
-    orientation: i32,
-) -> std::io::Result<()> {
-    let vf = build_image_thumb_filter(orientation, 200);
+/// Use ffmpeg to extract a 200px-wide thumbnail from formats the `image` crate
+/// can't decode (RAW: NEF/ARW, HEIC/HEIF). Writes JPEG bytes to `destination`
+/// regardless of its extension.
+pub fn generate_image_thumbnail_ffmpeg(path: &Path, destination: &Path) -> std::io::Result<()> {
     let output = Command::new("ffmpeg")
         .arg("-y")
         .arg("-i")
@@ -130,7 +101,7 @@ pub fn generate_image_thumbnail_ffmpeg(
         .arg("-vframes")
         .arg("1")
         .arg("-vf")
-        .arg(&vf)
+        .arg("scale=200:-1")
         .arg("-f")
         .arg("image2")
         .arg("-c:v")