Add Responses API support for reasoning models

[aymaneo]

Description

Add Responses API support for reasoning models

• Implement ResponsesAPIAdapter to normalize Responses API to ChatCompletion format
• Add helper functions for system to instructions extraction and tool format adaptation
• Introduce use_responses_api flag in Environment for opt-in
• Verified on AgentClinic Multiturn env with o3-mini and gpt-4o and all tests pass

Type of Change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• Test improvement

Testing

• All existing tests pass when running uv run pytest locally.
• New tests have been added to cover the changes

Checklist

• My code follows the style guidelines of this project as outlined in AGENTS.md
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• Any dependent changes have been merged and published

Additional Notes

[CLAassistant]

All committers have signed the CLA.

[willccbb]

thanks for the PR! adding proper support for responses + Anthropic APIs is on our roadmap, but we'll want to do it as part of a more holistic client abstraction which addresses a few other interrelated issues, rather than an if-else approach which feels a bit brittle/hard to maintain, so we'll probably hold off on merging this for now, sorry.