[AMD] perf: enable MiniMax M3 CUDA graphs on MI355X

[cquil11]

Remove --enforce-eager, set VLLM_USE_BREAKABLE_CUDAGRAPH=0, and disable prefix caching for the non-MTP MI355X recipe.

Validation: shell syntax and MI355X sweep generation.

---

Note

Low Risk
Benchmark-only launch flags for a single AMD recipe; no auth, serving API, or shared runtime code changes.

Overview
Aligns the MiniMax-M3 MXFP8 MI355X non-MTP vLLM recipe with the recent MI300X CUDA-graph change (#1750).

The benchmark script now exports VLLM_USE_BREAKABLE_CUDAGRAPH=0, drops --enforce-eager so CUDA graphs can run, and passes --no-enable-prefix-caching on vllm serve (consistent with other MiniMax random-dataset sweeps). perf-changelog.yaml documents the update for minimaxm3-fp8-mi355x-vllm.

^{Reviewed by Cursor Bugbot for commit 26bbb94. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

Thanks for the contribution! For vLLM & SGLang, please ensure that your recipes is similar to the official vLLM recipes and/or the SGLang cookbook

• https://docs.sglang.io/cookbook/intro
• https://recipes.vllm.ai/

If it is not, please create a PR first before we can merge your single node PR into the master branch. Let's ensure that the documentation is first class such that the entire ML community can benefit from your hard work! Thank you

• https://github.com/vllm-project/recipes
• https://github.com/sgl-project/sglang/tree/main/docs_new

PR authors are responsible for ensuring that after merging, all GitHub Action jobs fully pass. A lot of the time, failures are just flakes and simply re-running the failed jobs will fix it. If re-running failed jobs is attempted, PR authors are responsible for ensuring it passes. See GitHub's docs on re-running failed jobs: https://docs.github.com/en/actions/how-tos/manage-workflow-runs/re-run-workflows-and-jobs#re-running-failed-jobs-in-a-workflow

As a rule of thumb, generally, PR authors should request a review & get a PR approval from the respective companies' CODEOWNERS before requesting a review from core maintainers.

If additional help is needed, PR authors can reach out to core maintainers over Slack.

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=27491077319
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=27491077319

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=27491077523
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=27491077523

[cquil11]

/reuse-sweep-run

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=27500023263
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=27500023263