reasoningEffort overrides enableReasoningEffort setting in API config

[ikbencasdoei]

App Version

3.25.13

API Provider

OpenRouter

Model Used

claude-sonnet-4

Roo Code Task Links (Optional)

No response

🔁 Steps to Reproduce

1. Set "enableReasoningEffort": false in roo-code-settings.json → providerProfiles → apiConfigs → [profile-name].
2. Add "reasoningEffort": "medium".
3. Observe that the model still performs reasoning.

Environment:
Windows 10

💥 Outcome Summary

Expected reasoning to be disabled when enableReasoningEffort was false or missing, but got reasoning anyway when reasoningEffort: "medium" was present.

📄 Relevant Logs or Errors (Optional)

[roomote]

Hi! I'm here to help out the maintainers and am going to see if I can fix this issue. I'll investigate the reasoning effort configuration logic to ensure that when enableReasoningEffort is false, reasoning is properly disabled regardless of the reasoningEffort value.

[roomote]

I've successfully implemented a fix for this issue! 🎉

The problem was that the shouldUseReasoningEffort function didn't check the enableReasoningEffort setting. When enableReasoningEffort was set to false, reasoning would still be enabled if reasoningEffort was present.

Solution implemented:

• Modified shouldUseReasoningEffort in src/shared/api.ts to first check if enableReasoningEffort is explicitly set to false
• If it is false, the function now returns false immediately, properly disabling reasoning
• Added comprehensive test cases to verify the fix

Pull Request: #7049

All CI checks have passed successfully. The fix ensures that the enableReasoningEffort setting takes precedence over the reasoningEffort parameter, as expected.