[BUG] Mistral AI codestral-latest Context Window is incorrect

[machinekoder]

Problem (one or two sentences)

Expected: 128k token context window according to official docs
currently in Roo Code: 256k token limit

screenshot from official docs:

Context (who is affected and when)

Mistral Codestral Provioder

Reproduction steps

Create a Mistral Provider
Select Codestral-latest as Model

Expected result

128k context limit

Actual result

256k context limit

Variations tried (optional)

No response

App Version

latest

API Provider (optional)

Mistral AI

Model Used (optional)

codestral-latest

Roo Code Task Links (optional)

No response

Relevant logs or errors (optional)

[roomote]

I'm exploring a potential fix for the incorrect context window limit for Mistral's codestral-latest model and will open a PR to share progress.

[roomote]

Opened PR #9346 with my attempt at addressing this issue. The fix corrects the context window from 256k to 131k to match Mistral's official documentation. Feedback welcome!

[machinekoder]

Sounds good

[hannesrudolph]

@roomote make a new PR for this please