fix: guard against empty choices and message=None in LLM client

[qizwiz]

What

Add guards in LLMClient.complete() and LLMClient.complete_async() before accessing response.choices[0].message.content.

Why

chat.completions.create() can return two empty-response shapes, both unhandled:

1. choices = [] — on content-policy rejections, rate-limit errors, or provider failures
2. choices[0].message = None — Gemini 2.5 Flash (via the OpenAI-compatible endpoint) returns HTTP 200 with finish_reason: PROHIBITED_CONTENT and message=None rather than raising an exception

Both crash with IndexError or AttributeError. For a hallucination-prevention RAG system that may query multiple providers, silent crashes degrade the reliability guarantees the library is designed to provide.

Change

# Before
return response.choices[0].message.content

# After
if not response.choices or response.choices[0].message is None:
    raise ValueError("LLM returned empty or filtered response")
return response.choices[0].message.content

Applied to both complete() (sync) and complete_async() (async).

Corpus context

Detected by pact (llm_response_unguarded mode), a Z3-verified static analyzer for LLM crash vectors. This pattern was found across 13.7k violations in 786 repos.