[AMD] Port MI355X DeepSeek V4 fixed-sequence disagg to srt-slurm

[cquil11]

What changed

• ports the production MI355X DeepSeek V4 fixed-sequence STP and MTP disaggregated configurations to native srt-slurm orchestration
• preserves the TP8 concurrency grid (c1 through c128) and DEP8 grid (c256 and c512) inside four long-lived topology allocations
• uses the native SGLang Router frontend and AMD MoRI KV/MoE transport from the tested AMD runtime branch
• removes the superseded model-specific legacy submit wrapper

Runtime contract

• base: [AMD][AgentX] Port MI355X DeepSeek V4 MTP disagg to srt-slurm #2555
• srt-slurm: feat(runtime): add AMD accelerator support srt-slurm#1 at c609754b5622f96d5c12a93149e245308d4f1e9b
• container: lmsysorg/sglang-rocm:v0.5.15.post1-rocm720-mi35x-20260719
• benchmark: unchanged utils/bench_serving/benchmark_serving.py through the custom benchmark stage

Local validation

• all four srt-slurm dry-runs pass
• embedded benchmark commands pass bash -n
• generated matrix contains exactly four topology jobs
• pytest -q utils/matrix_logic: 231 passed
• git diff --check passes

Official MI355X full-sweep validation will be linked here and monitored through every topology/result.

[github-actions]

Thanks for the contribution! Please reach out to respective companies' CODEOWNER to fill in the latest PR_REVIEW_CHECKLIST.md before pinging core maintainer on Slack for review. In order for the signoff PR check bot to trigger, you must follow the PR_REVIEW_CHECKLIST.md template correctly, including the phrase As a PR reviewer and CODEOWNER, I have reviewed this and have.

For PR verification, add the full-sweep-fail-fast label (strongly recommended) to this PR — the benchmark sweep only runs on labeled PRs. Use full-sweep-enabled only if you need matrix jobs to keep running past a failure.

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. See GitHub's docs on re-running failed jobs

---

感谢你的贡献！请联系相应公司的 CODEOWNER 填写最新的 PR_REVIEW_CHECKLIST.md，然后再在 Slack 上联系核心维护者进行审阅。为了触发 signoff PR 检查机器人，你必须正确遵循 PR_REVIEW_CHECKLIST.md 模板，包括保留英文语句 As a PR reviewer and CODEOWNER, I have reviewed this and have。

如需进行 PR 验证，请为此 PR 添加 full-sweep-fail-fast 标签（强烈推荐）— 基准测试 sweep 仅在带有标签的 PR 上运行。仅当需要矩阵任务在失败后继续运行时才使用 full-sweep-enabled。

PR 作者有责任确保合并后所有 GitHub Action 任务完全通过。 很多时候失败只是偶发抖动（flake），重新运行失败的任务即可解决。参见 GitHub 关于重新运行失败任务的文档