fix: harden request handling, pipeline containment, and startup boundaries

backend/app/config.py

@@ -41,6 +41,8 @@ class Settings(BaseSettings):
     llm_model: str = "claude-sonnet-5"
     intent_effort: str = "low"
     rerank_effort: str = "medium"
+    # Bound provider calls independently from the grounding deadline.
+    llm_timeout_seconds: float = 120.0
     # Ceilings include adaptive thinking tokens, which is why they sit far
     # above the size of the structured output itself.
     intent_max_tokens: int = 16384