fix(openai-compat): inherit static thinking support

[shoucandanghehe]

Summary

• Inherit static model thinking support for OpenAI-compatible configured models before falling back to the generic low/medium/high default
• Include OpenAI-compatible model thinking metadata in the hot-reload model hash
• Add regressions for gpt-5.5 xhigh inheritance and thinking hash changes

Tests

• go test ./sdk/cliproxy ./internal/watcher/diff
• go build -o test-output ./cmd/server && rm test-output
• git diff --check

[github-actions]

This pull request targeted main.

The base branch has been automatically changed to dev.

[gemini-code-assist]

Code Review

This pull request updates the OpenAI compatibility model configuration to reuse static model definitions for thinking/reasoning capabilities when available, falling back to default levels if not. It also updates the model hash computation to include the thinking configuration and adds corresponding unit tests. The reviewer suggested avoiding a redundant lookup in buildOpenAICompatibilityConfigModels when modelID is identical to name.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: a262f58612

ℹ️ About Codex in GitHub

Codex has been enabled to automatically 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".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".