Add semantic note search (nomic-embed-text) and per-conversation note cache

- New NoteEmbedding model + migration 0014 stores float embeddings (JSONB)
- services/embeddings.py: get_embedding, upsert_note_embedding,
  semantic_search_notes (cosine similarity), backfill_note_embeddings
- build_context() now tries semantic search first, falls back to keyword search;
  accepts cached_note_ids to reuse last-turn notes and stabilise the system
  prompt prefix for Ollama's KV cache
- generation_buffer.py: per-conversation note ID cache (get/set/clear)
- generation_task.py: passes cached IDs into build_context, updates cache
  after each turn, and invalidates it after create_note/update_note/create_task
- app.py: pulls nomic-embed-text at startup and launches a background backfill
  to embed all existing notes (30 s delay so Ollama has time to load the model)
- routes/notes.py + services/tools.py: fire-and-forget embedding update on
  every note create or update via the API or LLM tool calls

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

src/fabledassistant/app.py

@@ -84,6 +84,7 @@ def create_app() -> Quart:
     async def startup():
         import asyncio
 
+        from fabledassistant.services.embeddings import backfill_note_embeddings
         from fabledassistant.services.generation_buffer import start_cleanup_loop
         from fabledassistant.services.llm import ensure_model
         from fabledassistant.services.logging import start_log_retention_loop
@@ -108,9 +109,22 @@ def create_app() -> Quart:
         models_to_pull = {Config.OLLAMA_MODEL}
         if Config.OLLAMA_INTENT_MODEL and Config.OLLAMA_INTENT_MODEL != Config.OLLAMA_MODEL:
             models_to_pull.add(Config.OLLAMA_INTENT_MODEL)
+        # Also pull the embedding model (nomic-embed-text by default).
+        models_to_pull.add(Config.EMBEDDING_MODEL)
         for _model in models_to_pull:
             asyncio.create_task(_pull_model(_model))
 
+        # After models are pulled, backfill embeddings for existing notes.
+        # Runs in the background so it never blocks the server from accepting requests.
+        async def _delayed_backfill() -> None:
+            await asyncio.sleep(30)  # Give Ollama time to load the embedding model
+            try:
+                await backfill_note_embeddings()
+            except Exception:
+                logger.warning("Embedding backfill failed", exc_info=True)
+
+        asyncio.create_task(_delayed_backfill())
+
     @app.route("/")
     async def serve_index():
         resp = await make_response(