fix: improve VS Code LM token usage reporting for context window updates

[roomote]

This PR fixes issue #6112 where the token count and context length progress bar don't update properly when using VS Code LM API as the API Provider.

The VS Code LM provider was only yielding token usage information once at the end of the stream, which prevented the UI from updating the context window progress bar during streaming.

Changes:

• Added initial usage yield with input tokens at stream start
• Added periodic token updates during streaming (every 500 chars)
• Included cache token fields for consistency with other providers

Fixes #6112

---

Important

Improves token usage reporting in VsCodeLmHandler by adding initial and periodic updates during streaming, addressing issue #6112.

• Behavior:
  • VsCodeLmHandler in vscode-lm.ts now yields initial token usage at stream start and periodic updates every 500 characters.
  • Final token usage is reported at stream end, with cache token fields set to 0.
• Tests:
  • Updated vscode-lm.spec.ts to expect initial, periodic, and final token usage chunks.
  • Added test for getApiProtocol in provider-settings.test.ts to return 'openai' for vscode-lm provider.
• Misc:
  • Fixes issue Token count or the Token progress bar doesn't work when using "VS Code LM API" as the API Provider #6112.

^{This description was created by for 048fcf7. You can customize this summary. It will automatically update as commits are pushed.}