fix: respect reasoning_config for custom providers and fix OpenRouter…

[Stephen30Yan]

… hardcoded reasoning

Two bugs fixed in _build_kwargs():

1. OpenRouter path (line 418): hardcoded extra_body.reasoning to {"enabled": True, "effort": "medium"} instead of reading the user's reasoning_config. Now respects the toggle switch.

2. Custom/unknown providers: reasoning_config was completely ignored because supports_reasoning was False for non-OpenRouter endpoints. Now sends top-level reasoning_effort param (which most OpenAI-compat APIs like DeepSeek/ARK expect). Maps "none" to "minimal" since some APIs reject "none" as invalid.

What does this PR do?

Related Issue

Fixes #

Type of Change

• 🐛 Bug fix (non-breaking change that fixes an issue)
• ✨ New feature (non-breaking change that adds functionality)
• 🔒 Security fix
• 📝 Documentation update
• ✅ Tests (adding or improving test coverage)
• ♻️ Refactor (no behavior change)
• 🎯 New skill (bundled or hub)

Changes Made

How to Test

Checklist

Code

• I've read the Contributing Guide
• My commit messages follow Conventional Commits (fix(scope):, feat(scope):, etc.)
• I searched for existing PRs to make sure this isn't a duplicate
• My PR contains only changes related to this fix/feature (no unrelated commits)
• I've run pytest tests/ -q and all tests pass
• I've added tests for my changes (required for bug fixes, strongly encouraged for features)
• I've tested on my platform:

Documentation & Housekeeping

• I've updated relevant documentation (README, docs/, docstrings) — or N/A
• I've updated cli-config.yaml.example if I added/changed config keys — or N/A
• I've updated CONTRIBUTING.md or AGENTS.md if I changed architecture or workflows — or N/A
• I've considered cross-platform impact (Windows, macOS) per the compatibility guide — or N/A
• I've updated tool descriptions/schemas if I changed tool behavior — or N/A

For New Skills

• This skill is broadly useful to most users (if bundled) — see Contributing Guide
• SKILL.md follows the standard format (frontmatter, trigger conditions, steps, pitfalls)
• No external dependencies that aren't already available (prefer stdlib, curl, existing Hermes tools)
• I've tested the skill end-to-end: hermes --toolsets skills -q "Use the X skill to do Y"

Screenshots / Logs

[alt-glitch]

Duplicate of #20594, which fixes the same two bugs (OpenRouter hardcoded reasoning effort ignoring reasoning_config, and custom/non-OpenRouter providers being excluded from reasoning forwarding). #20594 is the earlier open PR; consider consolidating there.

[teknium1]

Thanks for identifying the reasoning-config forwarding gap. This is now implemented on main through the live provider-profile path, so this older legacy-path patch is redundant.

• Automated hermes-sweeper review verified agent/chat_completion_helpers.py:947-986: registered providers return through provider_profile before the legacy fallback and receive reasoning_config.
• plugins/model-providers/custom/__init__.py:52-60 forwards configured effort as top-level reasoning_effort and emits think: false for disabled reasoning.
• plugins/model-providers/openrouter/__init__.py:149-160 forwards the supplied reasoning_config as OpenRouter reasoning data, with the necessary mandatory-Anthropic exception.
• Commit 67df958dbe06bb10ca16b8686c76baf0de3bac03 added the live custom-provider implementation and regression coverage; it shipped in v2026.7.7.
• The earlier duplicate discussion on fix: forward reasoning_config to custom providers (vLLM, Ollama, etc.) #20594 records the same live-path resolution and is now closed as implemented on main.