Update MG README with offgraph cache and batching

[metascroy]

This updates the MG README to reflect the offgraph cache and continuous batching support.

I'll hold off merging until the instructions are vetted for CUDA (cc @Gasoonjia)

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/23909

• 📄 Preview Python docs built from this PR

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

❌ You can merge normally! (1 Unrelated Failure), 1 Unclassified Failure

As of commit ddc40b7 with merge base 34b934b ():

UNCLASSIFIED FAILURE - DrCI could not classify the following job because the workflow did not run on the merge base. The failure may be pre-existing on trunk or introduced by this PR:

• MLX / test-mlx-llm (unsloth/Llama-3.2-1B-Instruct, llama-1b, false, nvfp4, macos-14-xlarge) / test-mlx-llm-llama-1b-nvfp4 (gh) (this job did not run on the merge base, so DrCI cannot tell whether the failure is pre-existing)
  RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 1

FLAKY - The following job failed but was likely due to flakiness present on trunk:

• MLX / test-mlx-nemotron3-diarization-build / test-mlx-nemotron3-diarization-build (gh) (detected as infra flaky with no log or failing log classifier)

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

[metascroy]

cc @Gasoonjia look over content for CUDA as well

[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.

[metascroy]

I think we can drop this. It doesn't matter what sampling params apply to the draft (it's an implementation detail).

Rejection sampling guarantees that the overall draft-target process has the sampling params requested by the user.