feat(api): update API spec from langfuse/langfuse 1ea859c

[langfuse-bot]

Greptile Summary

The generated API update expands LLM-as-judge prompts to ordered system, user, and assistant messages and publicly exports the new role enum.

• Adds EvaluatorChatMessageRole with system, user, and assistant values.
• Updates evaluator prompt documentation and message typing for multi-message prompts.
• Preserves the generated lazy-export chain across the evaluator and top-level API packages.

Confidence Score: 4/5

The public model compatibility break should be fixed before merging so existing callers can continue omitting the default user role.

The expanded enum and export wiring are consistent, but changing role from a defaulted field to a required field makes previously valid EvaluatorChatMessage(content=...) construction fail locally.

Files Needing Attention: langfuse/api/evaluators/types/evaluator_chat_message.py

Prompt To Fix All With AI

### Issue 1
langfuse/api/evaluators/types/evaluator_chat_message.py:15
**Required role breaks existing callers**

When existing callers construct `EvaluatorChatMessage(content="...")`, making `role` required removes the previous implicit `"user"` value, causing Pydantic validation to fail before an evaluator create or update request can be sent.

```suggestion
    role: EvaluatorChatMessageRole = EvaluatorChatMessageRole.USER
```

---

For each issue above, determine whether it is valid and should be fixed. If so, fix it directly.

_{Reviews (1): Last reviewed commit: "feat(api): update API spec from langfuse..." | Re-trigger Greptile}

Greptile also left 1 inline comment on this PR.

Context used:

• Knowledge Base — API client and service surface
• Knowledge Base — Generated API service domains

[github-actions]

@claude review

[claude]

Code review found no issues

No high-confidence issues detected in this change.