fix: add GLM-4.6 reasoning support for OpenAI Compatible provider

[roomote]

This PR attempts to address Issue #9012. Feedback and guidance are welcome.

Problem

GLM-4.6 model reasoning was not working correctly in v3.30.0:

• Z AI provider's GLM-4.6 could not properly enable reasoning
• OpenAI Compatible provider's GLM-4.6 showed reasoning but didn't display reasoning content, with reasoning duration always showing 0

Solution

Added GLM-4.6-specific reasoning support by detecting when:

• The model ID includes "glm-4.6"
• Reasoning is enabled (enableReasoningEffort: true)
• The model supports binary reasoning (supportsReasoningBinary: true)

When these conditions are met, the thinking: { type: "enabled" } parameter is added to the API request, which is the vendor-specific format required by Z AI's GLM models.

Changes

• Added GLM-4.6 detection logic in the OpenAI handler
• Added thinking parameter for both streaming and non-streaming modes
• Added comprehensive test coverage for GLM-4.6 reasoning scenarios
• Added explanatory comment for future maintainers

Testing

• All existing tests pass
• Added 4 new test cases covering:
  • GLM-4.6 with reasoning enabled (streaming)
  • GLM-4.6 with reasoning disabled (streaming)
  • GLM-4.6 with reasoning enabled (non-streaming)
  • Non-GLM-4.6 models with reasoning enabled (should not add thinking parameter)

Fixes #9012

---

Important

Adds GLM-4.6 reasoning support to OpenAI Compatible provider by introducing thinking parameter in OpenAiHandler.

• Behavior:
  • Adds GLM-4.6 reasoning support in OpenAiHandler by checking for glm-4.6 in model ID, enableReasoningEffort: true, and supportsReasoningBinary: true.
  • Adds thinking: { type: "enabled" } parameter to API requests when conditions are met.
• Testing:
  • Adds 4 new test cases in openai.spec.ts for GLM-4.6 reasoning scenarios, covering both streaming and non-streaming modes.
  • Ensures thinking parameter is not added for non-GLM-4.6 models even with reasoning enabled.

^{This description was created by for c0d0e13. You can customize this summary. It will automatically update as commits are pushed.}

[roomote]

Review completed. Found 1 issue that should be addressed before merging.

• Fix potential parameter conflict between thinking and reasoning_effort parameters

_{Mention @roomote in a comment to trigger your PR Fixer agent and make changes to this pull request.}

[roomote]

Potential parameter conflict: The thinking parameter (added for GLM-4.6 on lines 178-180) could coexist with reasoning_effort from the spread reasoning object on this line. If a GLM-4.6 model has both supportsReasoningBinary: true and supportsReasoningEffort: true configured, both parameters would be sent to the API. Since the PR description states GLM-4.6 uses thinking instead of reasoning_effort, consider excluding the reasoning object when adding the thinking parameter to prevent both from being present simultaneously.

_{Fix it with Roo Code or mention @roomote and request a fix.}