feat: Add gpt-5.5 support

[domwst]

📝 Summary

Closes #

📋 Pre-Review Checklist

• For large changes, or changes that affect the public API: this change was discussed or approved through an issue, on Discord, or the community discussions (Please provide a link if applicable).
• Any AI generated code has been reviewed line-by-line by the human PR author, who stands by it.
• Video or media evidence is provided for any visual changes (optional).

✅ Merge Checklist

• I have read the contributor guidelines.
• Documentation has been updated where applicable, including docstrings for API changes.
• Tests have been added for the changes made.

Hopefully most of the checks are not applicable to this change :)

I've used this version locally, it seem to work just fine. As per thinking setting, now sure what the policy is --
previous openai models that support reasoning are not marked as such. Description is copied from the
official model card

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
marimo-docs	Ready	Preview, Comment	May 9, 2026 10:29am

[github-actions]

All contributors have signed the CLA ✍️ ✅
_{Posted by the CLA Assistant Lite bot.}

[cubic-dev-ai]

No issues found across 1 file

[domwst]

I have read the CLA Document and I hereby sign the CLA

[domwst]

According to my testing, marimo sends an ungodly amount of context to the api on every request -- context window quickly runs out and simple tasks cost ~0.5$ on gpt-5.5 (wich is way more than I would've expected). I'll post more details here if I find anything related to this issue in the implementation.

[Light2Dark]

Thanks @domwst , is that on the agent mode / manual mode on the chat sidebar?
Agent mode may be inefficient, we could probably improve with tool discovery.

[domwst]

Yeah, it was agent mode