feat(microsoft-foundry): update model YAMLs [bot]

[models-bot]

Auto-generated by poc-agent for provider microsoft-foundry.

---

Note

Low Risk
Catalog-only YAML changes; the DeepSeek limit reduction may affect routing or UX that assumed 384k outputs, but there is no application logic change.

Overview
Updates microsoft-foundry model catalog metadata for two models.

DeepSeek-V4-Pro lowers max_output_tokens and max_tokens from 384000 to 128000, so advertised completion limits match the provider’s current cap.

gpt-5.1-codex-max adds datazone_us and datazone_eu cost rows (input, output, and cache-read pricing) ahead of the existing region: "*" entry, consistent with other Foundry models’ regional pricing.

^{Reviewed by Cursor Bugbot for commit 456836c. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

/test-models

[cursor]

Cursor Bugbot has reviewed your changes using default effort and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 456836c. Configure here.}

[cursor]

Missing max_tokens override

Medium Severity

This fuller DeepSeek-V4-Pro entry sets limits.max_output_tokens to 128000, above the inherited microsoft-foundry default.yaml max_tokens cap of 4096, without a params override for max_tokens.maxValue. Callers inherit that 4096 cap and cannot use the model's 128k output window.

 

^{Triggered by learned rule: Override max_tokens maxValue to match limits}

^{Reviewed by Cursor Bugbot for commit 456836c. Configure here.}