fix: Fix reasoning_effort not being passed through to OpenAI Chat Completion API

[green-david]

Description

This is in the same vein as #244. While testing Groq, I noticed that reasoning_effort was not getting passed through and sent to the Chat Completions API for GPT OSS models. The ChatAPI module was retrieving reasoning_effort from provider options, but it is a global common option that appears to be handled almost exclusively as a top-level option, not a provider option.

Type of Change

• Bug fix (non-breaking change fixing an issue)
• New feature (non-breaking change adding functionality)
• Breaking change (fix or feature causing existing functionality to change)
• Documentation update

Breaking Changes

I don't think this is a breaking change as it didn't appear to work as documented in the first place.

Testing

• Tests pass (mix test)
• Quality checks pass (mix quality)

Checklist

• My code follows the project's style guidelines
• I have updated the documentation accordingly
• I have added tests that prove my fix/feature works
• All new and existing tests pass
• My commits follow conventional commit format
• I have NOT edited CHANGELOG.md (it is auto-generated by git_ops)

Related Issues

#244

[mikehostetler]

Thanks for the contribution here. The implementation is directionally right, the PR is merge-clean, and CI is green.

I marked this needs_work because the exact bug fix needs a focused regression test: top-level reasoning_effort should make it into the OpenAI Chat Completions request body, not only when nested under provider_options. A small provider/unit test around ChatAPI.encode_body/1 or OpenAI.prepare_request/4 for a chat-compatible model should be enough. Once that coverage is in, this should be in good shape.

[green-david]

@mikehostetler Thanks for taking a look. I force-pushed an update adding a test in test/providers/openai_test.exs. Let me know if this needs anything else!