[FEATURE]:Support full 1M context windows for models that officially provide it

Feature hasn't been suggested before.

• I have verified this feature I'm about to request hasn't been suggested before.

Describe the enhancement you want to request

Hi OpenCode team,

First of all, thank you for OpenCode Go and the great work you're doing.

I recently subscribed to OpenCode Go because several supported models, such as DeepSeek V4 Pro, GLM-5.2, and Qwen3.7 Max, officially support a 1M token context window.

However, OpenCode currently shows:

Auto-compact window: 190k tokens

I would like to kindly request support for using the full 1M context window on models that officially provide it. This would be a valuable improvement for users working with large codebases and long development sessions.

Could you please let me know:

• Is the 190k auto-compact window an intentional limitation?
• Is there a way to enable the full 1M context window?
• If not, are there any plans to support it in the future?

Thank you for your time and for considering this request.