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/services/generation_task.py

@@ -15,7 +15,13 @@ from sqlalchemy import update
 from fabledassistant.config import Config
 from fabledassistant.models import async_session
 from fabledassistant.models.conversation import Message
-from fabledassistant.services.generation_buffer import GenerationBuffer, GenerationState
+from fabledassistant.services.generation_buffer import (
+    GenerationBuffer,
+    GenerationState,
+    clear_conv_note_cache,
+    get_conv_note_cache,
+    set_conv_note_cache,
+)
 from fabledassistant.services.llm import ChatChunk, build_context, generate_completion, stream_chat, stream_chat_with_tools, summarize_history_for_context
 from fabledassistant.services.chat import update_conversation_title
 from fabledassistant.services.intent import IntentResult, classify_intent
@@ -196,6 +202,10 @@ async def run_generation(
         history_to_use, history_summary = await summarize_history_for_context(history, intent_model)
 
     # Phase 3: Build context and classify intent in parallel — the two slow legs.
+    # Pass cached note IDs so build_context can reuse them, keeping the system
+    # prompt prefix stable and helping Ollama's KV cache stay warm.
+    cached_note_ids = get_conv_note_cache(conv_id) or None
+
     pre_intent: IntentResult = IntentResult()
     intent_timing_ms: int | None = None
     if tools:
@@ -209,6 +219,7 @@ async def run_generation(
             user_id, history_to_use, context_note_id, user_content,
             exclude_note_ids=exclude_note_ids,
             history_summary=history_summary,
+            cached_note_ids=cached_note_ids,
         ))
         intent_task = asyncio.create_task(
             classify_intent(user_content, tools, intent_model, history=intent_history)
@@ -220,8 +231,14 @@ async def run_generation(
             user_id, history_to_use, context_note_id, user_content,
             exclude_note_ids=exclude_note_ids,
             history_summary=history_summary,
+            cached_note_ids=cached_note_ids,
         )
 
+    # Update the note cache with whatever notes ended up in context.
+    new_note_ids = context_meta.get("auto_note_ids") or []
+    if new_note_ids:
+        set_conv_note_cache(conv_id, new_note_ids)
+
     # Emit context event
     buf.append_event("context", {"context": context_meta})
 
@@ -332,6 +349,11 @@ async def run_generation(
                             if timing["ttft_ms"] is None:
                                 timing["ttft_ms"] = int((time.monotonic() - t_start) * 1000)
 
+                        # Invalidate the note context cache after any successful note write
+                        # so the next turn can pick up newly created/modified notes.
+                        if result.get("success") and tool_name in {"create_task", "create_note", "update_note"}:
+                            clear_conv_note_cache(conv_id)
+
                         tool_record = {
                             "function": tool_name,
                             "arguments": intent.arguments,
@@ -400,6 +422,9 @@ async def run_generation(
                         timing["tools"].append({"name": tool_name, "ms": int((time.monotonic() - t_tool) * 1000)})
                         logger.info("Tool %s result: success=%s", tool_name, result.get("success"))
 
+                        if result.get("success") and tool_name in {"create_task", "create_note", "update_note"}:
+                            clear_conv_note_cache(conv_id)
+
                         tool_record = {
                             "function": tool_name,
                             "arguments": arguments,