fix: resolve VS Code LM API model selection issue

[roomote]

This PR attempts to address Issue #8991. Feedback and guidance are welcome.

Problem

All models in VS Code LM API return a 400 error "Model is not supported for this request" when users try to use them.

Root Cause

The VS Code Language Model API requires exact model IDs for selection, but the implementation was using vendor/family pairs instead.

Solution

• Updated UI component to store and use model.id instead of vendor/family
• Modified VsCodeLmHandler to select models by ID when available
• Added fallback mechanism (ID → selector → default model)
• Improved error messages to guide users when model approval is needed

Changes

• UI: Updated dropdown to use model.id as value
• Handler: Added ID-based model selection with proper fallback
• Error Handling: Added specific messages for model approval failures
• Tests: Added coverage for ID-based model selection

Testing

• All tests pass (unit tests for both UI and handler)
• Type checking passes
• Linting passes

Fixes #8991

---

Important

Fixes VS Code LM API model selection by using model IDs, updates UI and error handling, and adds tests.

• Behavior:
  • Fixes model selection in VsCodeLmHandler by using model.id instead of vendor/family pairs.
  • Adds fallback mechanism for model selection (ID → selector → default model).
  • Improves error messages for model approval in vscode-lm.ts.
• UI:
  • Updates dropdown in VSCodeLM.tsx to use model.id as value.
• Tests:
  • Adds test for ID-based model selection in vscode-lm.spec.ts.
• Misc:
  • Updates getVsCodeLmModels to filter and map models correctly in vscode-lm.ts.

^{This description was created by for 2da41b4. You can customize this summary. It will automatically update as commits are pushed.}

[roomote]

Review completed. Found 1 issue that should be addressed.

• Fallback logic in createClient doesn't work correctly when model ID is not found (lines 126-133 in vscode-lm.ts)

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

[roomote]

The fallback logic here doesn't work as intended. When a model with the specified ID is not found in allModels, the code attempts to fall back by calling selectChatModels(selector), but selector still contains the unfound ID. This means the fallback will return an empty array (since we already confirmed the ID doesn't exist), making the fallback ineffective. To fix this, create a new selector without the ID for the fallback: const { id, ...fallbackSelector } = selector and then use selectChatModels(fallbackSelector) to search by vendor/family only.