feat: add Show Debug Thoughts option for LM Studio provider

[loject]

Related GitHub Issue

Closes: #8140

Roo Code Task Context (Optional)

Description

This PR adds a "Show Debug Thoughts" feature for the LM Studio provider to help users detect infinite reasoning loops in open-source models like oss-gpt-20b.

Key implementation details:

• Schema Extension: Added lmStudioShowDebugThoughts boolean field to the provider settings schema with proper Zod validation
• API Integration: Enhanced the LM Studio handler to capture and conditionally display reasoning output from the delta.reasoning field
• UI Component: Added a settings checkbox with descriptive label and help text in the LM Studio provider configuration panel
• Localization: Implemented translations for both English and Russian locales
• Type Safety: Maintained full TypeScript compatibility across all changes

Reviewers should pay attention to:

• The reasoning output handling logic in src/api/providers/lm-studio.ts lines 123-133
• Proper integration with existing XmlMatcher for <think> tags vs new reasoning field
• UI placement and styling consistency with other LM Studio settings

Test Procedure

Manual Testing Steps:

1. Open Roo Code settings and navigate to LM Studio provider configuration
2. Verify the "Show Debug Thoughts" checkbox appears with proper label and description
3. Toggle the checkbox and confirm the setting is saved
4. Test with a reasoning-capable model (if available) to verify reasoning output display
5. Verify localization by switching between English and Russian language settings

Automated Testing:

• All existing unit tests pass (1075/1075 tests successful)
• ESLint validation passes with no warnings
• TypeScript compilation succeeds without errors
• Settings schema validation works correctly

Pre-Submission Checklist

• Issue Linked: This PR is linked to an approved GitHub Issue (see "Related GitHub Issue" above).
• Scope: My changes are focused on the linked issue (one major feature/fix per PR).
• Self-Review: I have performed a thorough self-review of my code.
• Testing: New and/or updated tests have been added to cover my changes (if applicable).
• Documentation Impact: I have considered if my changes require documentation updates (see "Documentation Updates" section below).
• Contribution Guidelines: I have read and agree to the Contributor Guidelines.

Screenshots / Videos

Before: LM Studio settings without debug thoughts option
After: LM Studio settings with new "Show Debug Thoughts" checkbox and description

Screenshots will be provided once the PR is ready for submission

Documentation Updates

• No documentation updates are required.
• Yes, documentation updates are required.

Documentation needs:

• Update LM Studio provider documentation to mention the debug thoughts feature
• Add troubleshooting section about detecting infinite reasoning loops
• Include usage examples for debugging OSS model behavior

Additional Notes

• This feature is opt-in and disabled by default to avoid UI clutter
• The implementation differentiates between standard <think> tags (always processed) and the new reasoning field (conditionally displayed)
• Performance impact is minimal when the feature is disabled
• Compatible with existing LM Studio speculative decoding and draft model features

Get in Touch

Discord: loject

---

Important

Adds 'Show Debug Thoughts' feature to LM Studio provider, including schema, UI, and localization updates.

• Behavior:
  • Adds lmStudioShowDebugThoughts boolean field to provider settings schema in provider-settings.ts.
  • Updates LmStudioHandler in lm-studio.ts to display reasoning output when lmStudioShowDebugThoughts is enabled.
• UI:
  • Adds "Show Debug Thoughts" checkbox in LMStudio.tsx with label and description.
• Localization:
  • Adds translations for "Show Debug Thoughts" feature in multiple locale files, including settings.json for ru, tr, vi, zh-CN, and zh-TW.

^{This description was created by for 61b108f. You can customize this summary. It will automatically update as commits are pushed.}

[roomote]

Thank you for your contribution! I've reviewed the changes and found several issues that need attention before this can be merged.

[daniel-lxs]

Hey @loject, thank you for your contribution!

Does it make sense to just enable reasoning for LM Studio? We already show reasoning for every other provider; it makes sense to do it here.

[loject]

@daniel-lxs hi!
Just to clarify — here we’re talking about the debug reasoning traces, not the normal outputs. Some models (like gpt-oss-20b) don’t expose them in the standard stream, so it wouldn’t behave the same way as with the other providers.