[NV] Add Kimi K2.5 FP4 B200/B300 EP sweep

[jasonlizhengjian]

Add expert-parallel sweep points for Kimi K2.5 FP4 B200/B300 vLLM aggregate benchmarks.

Conditionally enables vLLM expert parallelism when EP_SIZE > 1.

---

Note

Low Risk
Benchmark matrix and launch-script wiring only; no production serving, auth, or data-path changes.

Overview
Extends Kimi K2.5 FP4 aggregate vLLM fixed-seq-len benchmarks on B200 and B300 by adding expert-parallel sweep points (ep: 8 with tp: 8, ep: 4 with tp: 4) for both 1k/1k and 8k/1k scenarios in .github/configs/nvidia-master.yaml, alongside the existing ep: 1 points.

The B200/B300 launch scripts now require EP_SIZE and pass --enable-expert-parallel to vllm serve only when EP_SIZE > 1. perf-changelog.yaml records the new sweep for kimik2.5-fp4-b200-vllm and kimik2.5-fp4-b300-vllm.

^{Reviewed by Cursor Bugbot for commit 8c2f793. 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=26902572165
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=26902572165

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=26902572165
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=26902572165