Add sleep/wake support for diffusion engine

[MikukuOvO]

Motivation

This PR is inherited from and rebased on top of #19152

It adds coarse-grained sleep/wake support for the multimodal diffusion engine so the server can temporarily release GPU memory without restarting the process.

Modifications

• Add /release_memory_occupation and /resume_memory_occupation for diffusion.
• Add scheduler/worker support to move active pipeline modules between GPU and CPU.
• Reject generation requests while the server is sleeping.
• Add tests and documentation for the sleep/wake flow.

Accuracy Tests

This PR does not change model math or kernel behavior.

Functional checks were run for:

• Qwen/Qwen-Image on 1 GPU
• Qwen/Qwen-Image with TP=2
• Tongyi-MAI/Z-Image
• Tongyi-MAI/Z-Image-Turbo

Verified behavior:

• sleep releases GPU memory
• generation while sleeping returns 400
• wake restores service and generation works again

Speed Tests and Profiling

This PR is not intended as an inference speed optimization.

Checklist

• Format your code according to the [Format code with pre-commit](https://docs.sglang.io/developer_guide/
  contribution_guide.html#format-code-with-pre-commit).
• Add unit tests according to the [Run and add unit tests](https://docs.sglang.io/developer_guide/contribution_guide.html#run-
  and-add-unit-tests).
• Update documentation according to [Write documentations](https://docs.sglang.io/developer_guide/contribution_guide.html#write-
  documentations).
• Provide accuracy and speed benchmark results according to [Test the accuracy](https://docs.sglang.io/developer_guide/
  contribution_guide.html#test-the-accuracy) and [Benchmark the speed](https://docs.sglang.io/developer_guide/
  contribution_guide.html#benchmark-the-speed).
• Follow the SGLang code style guidance.

Review and Merge Process

1. Ping Merge Oncalls to start the process. See the [PR Merge Process](https://github.com/sgl-project/sglang/blob/main/.github/
   MAINTAINER.md#pull-request-merge-process).
2. Get approvals from CODEOWNERS and other reviewers.
3. Trigger CI tests with comments or
   contact authorized users to do so.
   • Common commands include /tag-and-rerun-ci, /tag-run-ci-label, /rerun-failed-ci
4. After green CI and required approvals, ask Merge Oncalls or people with Write permission to merge the PR.

[gemini-code-assist]

Warning

You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again!