[executorch][muse-glimmer] Add CUDA speculative sampling orchestration

[Gasoonjia]

Stack from ghstack (oldest at bottom):

• [executorch][muse-glimmer] Run DFlash sampling on CUDA #22492
• -> [executorch][muse-glimmer] Add CUDA speculative sampling orchestration #22491
• [executorch][muse-glimmer] Add unified CUDA sample_token entry point #22490
• [executorch][muse-glimmer] Add CUDA residual correction sampling #22489
• [executorch][muse-glimmer] Add CUDA excluded-token correction #22488
• [executorch][muse-glimmer] Add CUDA speculative acceptance #22487
• [executorch][muse-glimmer] Add CUDA categorical sampling #22486
• [executorch][muse-glimmer] Add CUDA sampling probabilities #22485
• [executorch][muse-glimmer] Add CUDA argmax sampling primitive #22484

Add the CUDA counterparts of the DFlash verification step. stochastic_speculative_sample walks the candidate block on device using already normalized target and draft probability rows, applies the host acceptance rule, and falls back to residual or excluded-token correction on the first rejection, while greedy_speculative_sample compares committed argmax tokens without touching probabilities. Both return the committed candidate count and correction token as device-resident scalars so the decode loop never needs the full probability rows on the host.

Differential Revision: D117826289

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/22491

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❌ 1 New Failure

As of commit 4a46bb3 with merge base c65ad53 ():

NEW FAILURE - The following job has failed:

• Lint / lintrunner (gh)
  >>> Lint for examples/models/muse-glimmer/tests/sampling_cuda_test.cpp:

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[github-actions]

This PR needs a release notes: label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.