feat: add diagnostic logging for GLM model detection

[roomote]

This PR attempts to address Issue #11071 by adding diagnostic logging to help users verify GLM model detection.

Problem

Users testing PR #11077 cannot verify whether:

• Their model ID is being correctly identified as a GLM model
• The GLM-specific optimizations are being applied
• They are using the correct custom build vs. marketplace version

Solution

Added console logging at key points:

1. Model ID logging: Logs the model ID being used when LM Studio or OpenAI-compatible providers initialize
2. GLM detection results: Logs when a GLM model is detected with specific optimization settings
3. Applied optimization values: Logs the parallel_tool_calls value being sent to the API

Example Output

When using a GLM model like glm-4.5:

[LM Studio] Using model ID: "glm-4.5"
[GLM Detection] ✓ GLM model detected: "glm-4.5"
[GLM Detection]   - mergeToolResultText: true
[GLM Detection]   - disableParallelToolCalls: true
[LM Studio] parallel_tool_calls set to: false

When using a non-GLM model:

[LM Studio] Using model ID: "llama-3-8b"
[LM Studio] parallel_tool_calls set to: true

Testing

• ✅ All existing tests pass
• ✅ Type checking passes
• ✅ Linting passes

Notes

• This PR builds on top of PR feat: add GLM model detection for LM Studio and OpenAI-compatible providers #11077
• The logging helps diagnose why the file read looping may persist even with GLM detection
• Users can check VS Code Developer Tools console to see these logs

Feedback and guidance are welcome!

---

Important

Adds diagnostic logging for GLM model detection and optimizations, with utility functions and tests for model identification and option retrieval.

• Behavior:
  • Adds diagnostic logging for GLM model detection in base-openai-compatible-provider.ts and lm-studio.ts.
  • Logs model ID, GLM detection results, and parallel_tool_calls value.
• Utilities:
  • Adds isGlmModel() and getGlmModelOptions() in model-detection.ts for GLM model detection and option retrieval.
• Testing:
  • Adds tests in model-detection.spec.ts for GLM model detection and option retrieval.

^{This description was created by for 2c5d905. You can customize this summary. It will automatically update as commits are pushed.}

[roomote]

Rooviewer   See task on Roo Cloud

Review complete. No issues found.

The implementation correctly:

• Detects GLM models using pattern matching (glm-, glm4, chatglm)
• Applies appropriate options (mergeToolResultText, disableParallelToolCalls)
• Integrates with both base-openai-compatible-provider.ts and lm-studio.ts
• Has comprehensive test coverage for model detection edge cases

Diagnostic logging output matches the expected format from the PR description.

_{Mention @roomote in a comment to request specific changes to this pull request or fix all unresolved issues.}