Speed up: preload BM25 at startup, httpx timeout+limits, fix lambda in to_thread

- index: lifespan preloads BM25 model so first /sparse_embedding request
  doesn't pay cold-start cost (~1-2s per worker)
- search: same BM25 preload + httpx timeout=30s + connection limits to
  avoid hanging on slow external APIs
- search: asyncio.to_thread(fn, arg) instead of lambda wrapper

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

index/main.py

@@ -1,6 +1,7 @@
 import asyncio
 import logging
 import os
+from contextlib import asynccontextmanager
 from functools import lru_cache
 from typing import Any
 
@@ -189,7 +190,13 @@ def build_chunks(
     return result
 
 
-app = FastAPI(title="Index Service", version="0.1.0")
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    await asyncio.to_thread(get_sparse_model)
+    yield
+
+
+app = FastAPI(title="Index Service", version="0.1.0", lifespan=lifespan)
 
 
 @app.get("/health")

search/main.py

@@ -148,11 +148,10 @@ def get_sparse_model() -> SparseTextEmbedding:
 
 @asynccontextmanager
 async def lifespan(app: FastAPI):
-    app.state.http = httpx.AsyncClient()
+    await asyncio.to_thread(get_sparse_model)
-    app.state.qdrant = AsyncQdrantClient(
+    limits = httpx.Limits(max_connections=100, max_keepalive_connections=20)
-        url=QDRANT_URL,
+    app.state.http = httpx.AsyncClient(timeout=30.0, limits=limits)
-        api_key=API_KEY,
+    app.state.qdrant = AsyncQdrantClient(url=QDRANT_URL, api_key=API_KEY)
-    )
     try:
         yield
     finally:
@@ -354,7 +353,7 @@ async def search(payload: SearchAPIRequest) -> SearchAPIResponse:
     sparse_query = build_sparse_query(question)
 
     dense_task = embed_dense(client, dense_query)
-    sparse_task = asyncio.to_thread(lambda: embed_sparse_sync(sparse_query))
+    sparse_task = asyncio.to_thread(embed_sparse_sync, sparse_query)
     dense_vector, sparse_vector = await asyncio.gather(dense_task, sparse_task)
 
     dense_vectors = [dense_vector]