feat(ai): chat continuation for photo insights (server v1)

Adds POST /insights/chat and GET /insights/chat/history. Replays the
stored agentic conversation through the same backend the insight was
generated with (or a per-turn override), runs a short tool-calling
loop, and persists the extended history in append or amend mode.

Backend switching: same-backend or hybrid->local replay verbatim;
local->hybrid is rejected in v1 (would require on-the-fly vision
description rewrite).

Per-(library, file) async mutex serialises concurrent turns. Soft
context budget drops oldest tool_call+result pairs when the
serialized history exceeds num_ctx - 2048 tokens.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

src/ai/insight_generator.rs

@@ -96,7 +96,7 @@ impl InsightGenerator {
     /// first root under which the file exists. Insights may be generated
     /// for any library — the generator itself doesn't know which — so we
     /// probe each root rather than trust a single `base_path`.
-    fn resolve_full_path(&self, rel_path: &str) -> Option<std::path::PathBuf> {
+    pub(crate) fn resolve_full_path(&self, rel_path: &str) -> Option<std::path::PathBuf> {
         use std::path::Path;
         for lib in &self.libraries {
             let candidate = Path::new(&lib.root_path).join(rel_path);
@@ -129,7 +129,7 @@ impl InsightGenerator {
 
     /// Load image file, resize it, and encode as base64 for vision models
     /// Resizes to max 1024px on longest edge to reduce context usage
-    fn load_image_as_base64(&self, file_path: &str) -> Result<String> {
+    pub(crate) fn load_image_as_base64(&self, file_path: &str) -> Result<String> {
         use image::imageops::FilterType;
 
         let full_path = self.resolve_full_path(file_path).ok_or_else(|| {
@@ -1411,7 +1411,7 @@ Return ONLY the summary, nothing else."#,
     // ── Tool executors for agentic loop ────────────────────────────────
 
     /// Dispatch a tool call to the appropriate executor
-    async fn execute_tool(
+    pub(crate) async fn execute_tool(
         &self,
         tool_name: &str,
         arguments: &serde_json::Value,
@@ -2136,7 +2136,7 @@ Return ONLY the summary, nothing else."#,
     // ── Agentic insight generation ──────────────────────────────────────
 
     /// Build the list of tool definitions for the agentic loop
-    fn build_tool_definitions(has_vision: bool) -> Vec<Tool> {
+    pub(crate) fn build_tool_definitions(has_vision: bool) -> Vec<Tool> {
         let mut tools = vec![
             Tool::function(
                 "search_rag",