feat(openai): pass through code_interpreter tool in Responses API

[OmarGoubail]

Description

Adds minimal Code Interpreter passthrough support to ReqLLM's OpenAI Responses API provider.

Callers can now pass %{"type" => "code_interpreter", "container" => ...} in tools: and ReqLLM will forward the tool map unchanged to OpenAI.

Raw code_interpreter_* output items (code_interpreter_call, code_interpreter_logs, code_interpreter_interpretation, etc.) can be accessed in response.provider_meta["code_interpreter"]["items"] and are excluded from normal text/function tool-call extraction.

Both object containers (%{"type" => "auto", "memory_limit" => "4g"}) and explicit string container IDs ("cntr_abc123") are supported.

Notes:

• This is Responses API only. OpenAI's Chat Completions SDK type (ChatCompletionFunctionToolParam) only supports type: "function", so code_interpreter is not supported there.
• Azure's Responses API provider delegates to ReqLLM.Providers.OpenAI.ResponsesAPI, so this change will pass
  code_interpreter through automatically. However, Azure OpenAI's Responses API may not expose Code Interpreter; I am unable to confirm this through the docs though. I am working on trying to test it.

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

Testing

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

Live verification:

Screen.Recording.2026-06-23.at.3.16.58.PM.mov

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)