[jihtsan]

Decode OpenRouter's reasoning response field into the existing reasoningContent property for streamed deltas and nonstreaming chat messages. Keep reasoning_content precedence when both keys are present, including empty strings, and preserve the other response fields and synthesized memberwise initializers. Closes #168.

This handles the documented Chat Completions reasoning field; it does not change provider request settings. Related #175 changes request parameters/audio and leaves these response models unchanged.

Validation on macOS arm64 / Swift 6.4:

• Both OpenRouter response fixtures fail before the fix (reasoningContent == nil), then pass.
• Seven regression tests cover stream/nonstream text, legacy precedence, null/missing/empty values, UTF-8, existing tool/refusal/audio fields and invalid types.
• swift test --disable-automatic-resolution: 116 XCTest tests passed, 0 failures.
• swiftformat --lint --config rules.swiftformat <three changed files>: 0 files need formatting; git diff --check: passed.
• Independent Standards and Spec reviews: no findings, tied to this commit's source blobs.

No live OpenRouter/Gemini call was made; these fixtures verify decoding of reasoning when the provider returns it. iOS/Linux/Windows were not run locally.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-11-05T06:39:03.737858Z	6971e17	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.