[Feature Request] Support for alternative LLM providers beyond Anthropic Claude models

Question

**Bug Description**
gpt-5.5 thinks and reasons much better. i am reviewing all the implementations plans made by opus-4.7 at least 10 times with gpt-5.5 and I find a new issue with opus-4.7 plan every single time

**Environment Info**
- Platform: linux
- Terminal: vte-based
- Version: 2.1.143
- Feedback ID: 1453189b-9f56-4fb0-8e0d-0173b75b6fa3

**Errors**
```json
[]
```