What does this PR do?

Fixes the provider-profile chat_completions path so non-vision models still run Hermes' shared image fallback before transport kwargs are built. Without this, /v1/chat/completions can forward raw image_url parts to non-vision models and trigger 400s even though the legacy path already strips or describes those images.

Related Issue

Fixes #23733

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

• updated /run_agent.py so the provider-profile chat_completions branch calls _prepare_messages_for_non_vision_model(...) before build_kwargs(...)
• added a regression test in /tests/run_agent/test_run_agent.py that proves the provider-profile path now forwards transformed non-vision messages instead of raw multimodal payloads
• re-ran /tests/gateway/test_api_server_multimodal.py to verify the API server multimodal path still behaves correctly end to end

How to Test

1. uv run --frozen ruff check run_agent.py tests/run_agent/test_run_agent.py
2. uv run --frozen pytest -q -o addopts='' tests/run_agent/test_run_agent.py -k 'provider_profile_applies_non_vision_fallback or converts_multimodal_user_image_to_text or reuses_cached_image_analysis_for_duplicate_images'
3. uv run --frozen pytest -q -o addopts='' tests/gateway/test_api_server_multimodal.py

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: macOS 15.5

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

Screenshots / Logs

• Targeted regression pytest: 3 passed, 321 deselected
• API server multimodal suite: 23 passed
• Clean-env collect-only smoke: 347 tests collected