feat: v4.1.1 - Combined PRs #62, #63, #64

[numman-ali]

Summary

This release combines fixes and features from three community PRs into a single v4.1.1 release:

• PR feat: Add "none" reasoning effort support for GPT-5.2 and GPT-5.1 #62 (@ben-vargas): Add "none" reasoning effort support for GPT-5.2 and GPT-5.1
• PR fix: filter orphaned function_call_output even when tools present #63 (@code-yeongyu): Fix orphaned function_call_output 400 errors
• PR Update OAuth Authentication version in HTML #64 (@kanemontreuil): Update OAuth HTML version display

Changes

Added

• "none" reasoning effort support: GPT-5.1 and GPT-5.2 general purpose models now support reasoning_effort: "none" which disables the reasoning phase entirely for faster responses
  • gpt-5.2-none - GPT-5.2 with reasoning disabled
  • gpt-5.1-none - GPT-5.1 with reasoning disabled
• Model mappings: Added gpt-5.2-none and gpt-5.1-none to model-map.ts
• Model presets: Added none presets to opencode.json configs (now 18 total models)
• 4 new unit tests for "none" reasoning behavior (197 total tests)

Fixed

• Orphaned function_call_output 400 errors: Fixed API errors when conversation history contains item_reference pointing to stored function calls
  • Previously, orphaned function_call_output items were only filtered when !body.tools
  • Now always handles orphans regardless of tools presence
  • Converts orphans to assistant messages instead of removing them, preserving context while avoiding API errors
• OAuth HTML version display: Updated version in oauth-success.html from 1.0.4 to 4.1.0

Technical Details

• getReasoningConfig() now detects GPT-5.1 general purpose models (not Codex variants) and allows "none" to pass through
• GPT-5.2 inherits "none" support as it's newer than GPT-5.1
• Codex variants do NOT support "none":
  • Codex and Codex Max: "none" auto-converts to "low"
  • Codex Mini: "none" auto-converts to "medium"

Documentation Updated

• README.md: Updated to 18 models, updated reasoning effort table, version references
• docs/configuration.md: Updated reasoning effort documentation with complete support matrix
• scripts/test-all-models.sh: Added gpt-5.2-none and gpt-5.1-none tests
• CHANGELOG.md: Combined release notes for all 3 PRs

Reasoning Effort Support Matrix

Model	none	low	medium	high	xhigh
GPT-5.2	✅	✅	✅	✅	✅
GPT-5.1	✅	✅	✅	✅	❌
GPT-5.1-Codex	❌→low	✅	✅	✅	❌
GPT-5.1-Codex-Max	❌→low	✅	✅	✅	✅
GPT-5.1-Codex-Mini	❌→medium	❌→medium	✅	✅	❌→high

Test Plan

• TypeScript type checking passes
• All 197 unit tests pass (was 193, added 4 new tests)
• Build succeeds
• Manual testing with opencode using local dist

Related PRs

Supersedes:

• feat: Add "none" reasoning effort support for GPT-5.2 and GPT-5.1 #62 - feat: Add "none" reasoning effort support for GPT-5.2 and GPT-5.1
• fix: filter orphaned function_call_output even when tools present #63 - fix: filter orphaned function_call_output even when tools present
• Update OAuth Authentication version in HTML #64 - Update OAuth Authentication version in HTML

---

🤖 Generated with Claude Code