fix(chat): resolve Auto-route failover event-loop freeze (#812)

Root cause: ensure_structured_column() executed a synchronous blocking
engine.connect().execute(SELECT information_schema.columns ...) inside the
async request path (_build_chat_history / _maybe_compress_history), freezing
the uvicorn event loop and hanging all endpoints (~200s, recovered only by
docker restart).

Fix:
- chat_context.py: remove per-request ensure_structured_column() calls; run
  once idempotently at app startup (main.py lifespan). Use CURRENT_DATABASE()
  for PostgreSQL compatibility.
- main.py: add startup-only ensure_structured_column() hook (try/except).
- llm_client.py: trust_env=False + bounded httpx.Timeout/wait_for so broken
  endpoints fail fast instead of hanging the event loop.
- chat_service.py: run media generation in a background daemon thread with its
  own event loop (_launch_media_task_bg) so SSE end does not cancel image/video
  tasks (#811); remove TEMP faulthandler diagnostic.

Safe-align: persist accumulated live container changes (admin/*, knowledge,
market, models, config, observability, mcp_client, platform_knowledge,
im_bot_platform, chat_media, capability_registry, requirements) and add
workflow_templates.

Verified: live failover test (model-006 -> conn-refused) returns in ~1.9s with
concurrent login 0.83s/200; event-loop responsiveness restored (sub-second).

backend/python/main.py

@@ -118,6 +118,18 @@ def _record_api_call(db, method: str, path: str, status_code: int, elapsed: floa
 async def lifespan(app: FastAPI):
     logger.info("Qihu-AIFlow v2.0.0 启动中...")
 
+    # 幂等确保 conversation_summaries.structured_memory 列。
+    # 原实现在每次聊天的请求路径中同步执行 engine.connect().execute(SELECT information_schema.columns)，
+    # 该同步阻塞调用运行在 asyncio 事件循环线程上，会导致整个事件循环冻结（所有端点挂死，
+    # 表现为「单 worker 永久阻塞、docker restart 才恢复」）。改为启动时一次性执行，
+    # 既保留对全新库的兼容，又绝不阻塞任何用户请求。
+    try:
+        from app.api.chat_context import ensure_structured_column
+        ensure_structured_column()
+        logger.info("[startup] ensure_structured_column 完成")
+    except Exception as _e:
+        logger.warning(f"[startup] ensure_structured_column 失败(可忽略): {_e}")
+
     # 启动后台 Evolution Pipeline（知识库图谱已存在的用户配置自动恢复运行）
     try:
         from app.services.evolution_service import EvolutionWatcherManager