knowledge: stamp model + backend on facts for audit

Adds two nullable TEXT columns to entity_facts —
`created_by_model` (LLM identifier) and `created_by_backend`
("local" / "hybrid" / "manual" / NULL) — so the curator can audit
which configurations produce good fact-keeping and which produce
noise.

photo_insights already carries model_version + backend, and
entity_facts.source_insight_id links to it, but:
  - source_insight_id is set post-loop, so chat-continuation and
    regenerated-insight facts lose the link.
  - JOINing per read is more friction than embedding provenance on
    the row itself.
  - Manual facts (POST /knowledge/facts) have no insight at all and
    need their own "manual" provenance marker.

Threading: execute_tool grows `model` + `backend` params, passed
from the three call sites (agentic insight loop, chat single-turn,
chat stream) using the loop-time `chat_backend.primary_model()` +
`effective_backend` already in scope. tool_store_fact stamps the
new fact accordingly; manual create_fact stamps backend="manual".
Legacy rows leave both NULL — pre-tracking data can't be back-
filled reliably from training_messages without burning compute.

Indexes are partial (WHERE NOT NULL) so legacy rows don't bloat
them, and "show me all facts from model X" stays fast.

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

src/ai/insight_chat.rs

@@ -497,6 +497,8 @@ impl InsightChatService {
                             &normalized,
                             req.user_id,
                             &active_persona,
+                            &model_used,
+                            &effective_backend,
                             &loop_cx,
                         )
                         .await;
@@ -870,6 +872,8 @@ impl InsightChatService {
                 &normalized,
                 req.user_id,
                 &active_persona,
+                &model_used,
+                &effective_backend,
                 max_iterations,
                 &tx,
             )
@@ -1059,6 +1063,8 @@ impl InsightChatService {
                 &normalized,
                 req.user_id,
                 &active_persona,
+                &model_used,
+                &effective_backend,
                 max_iterations,
                 &tx,
             )
@@ -1210,6 +1216,10 @@ impl InsightChatService {
         normalized: &str,
         user_id: i32,
         active_persona: &str,
+        // Provenance — stamped onto any store_fact tool call made
+        // during this loop. Mirrors the non-streaming chat path.
+        model_used: &str,
+        effective_backend: &str,
         max_iterations: usize,
         tx: &tokio::sync::mpsc::Sender<ChatStreamEvent>,
     ) -> Result<AgenticLoopOutcome> {
@@ -1290,6 +1300,8 @@ impl InsightChatService {
                             normalized,
                             user_id,
                             active_persona,
+                            model_used,
+                            effective_backend,
                             &cx,
                         )
                         .await;

src/ai/insight_generator.rs

@@ -1554,6 +1554,13 @@ Return ONLY the summary, nothing else."#,
         file_path: &str,
         user_id: i32,
         persona_id: &str,
+        // Provenance — written into entity_facts.created_by_* when
+        // the loop calls store_fact. The caller knows the actual
+        // chat-runtime model and backend (which may differ from
+        // ollama.primary_model in hybrid mode where chat lives on
+        // OpenRouter while Ollama still handles vision).
+        model: &str,
+        backend: &str,
         cx: &opentelemetry::Context,
     ) -> String {
         let result = match tool_name {
@@ -1574,8 +1581,10 @@ Return ONLY the summary, nothing else."#,
             }
             "store_entity" => self.tool_store_entity(arguments, ollama, cx).await,
             "store_fact" => {
-                self.tool_store_fact(arguments, file_path, user_id, persona_id, cx)
-                    .await
+                self.tool_store_fact(
+                    arguments, file_path, user_id, persona_id, model, backend, cx,
+                )
+                .await
             }
             "get_current_datetime" => Self::tool_get_current_datetime(),
             unknown => format!("Unknown tool: {}", unknown),
@@ -2632,6 +2641,8 @@ Return ONLY the summary, nothing else."#,
         file_path: &str,
         user_id: i32,
         persona_id: &str,
+        model: &str,
+        backend: &str,
         cx: &opentelemetry::Context,
     ) -> String {
         use crate::database::models::{InsertEntityFact, InsertEntityPhotoLink};
@@ -2700,6 +2711,8 @@ Return ONLY the summary, nothing else."#,
             valid_from,
             valid_until: None,
             superseded_by: None,
+            created_by_model: Some(model.to_string()),
+            created_by_backend: Some(backend.to_string()),
         };
 
         let mut kdao = self
@@ -3730,6 +3743,8 @@ Return ONLY the summary, nothing else."#,
                             &file_path,
                             user_id,
                             &persona_id,
+                            chat_backend.primary_model(),
+                            &backend_label,
                             &loop_cx,
                         )
                         .await;