fix: guard against empty choices and message=None across all model backends

[qizwiz]

What

Add explicit guards in all four model backend classes before accessing response.choices[0].message.content.

Why

client.chat.completions.create() can return two empty-response shapes, neither currently handled:

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

Both crash with IndexError or AttributeError. All four model backends have identical unguarded access patterns.

Files changed

File	Fix
common/models/openai_model.py	Guard before choices[0].message.content
common/models/azure_model.py	Guard before choices[0].message.content
common/models/cerebras_model.py	Guard before choices[0].message.content
common/models/groq_model.py	Guard before choices[0].message.content

# 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

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.