cbc9eeb453
- Add Ollama to allowed providers in model list endpoint - Add model pattern detection for Ollama models (glm-, qwen, gemma, llama, mistral, codellama) - This fixes 500 errors when using Ollama models via /v1/chat/completions