[executorch][muse] Add run_solo_batching

[Gasoonjia]

Stack from ghstack (oldest at bottom):

• [executorch][muse] Move off-graph KV to the batching path and test batching in OSS CI #23375
• -> [executorch][muse] Add run_solo_batching #23374
• [executorch][muse] Add export_solo_batching.py #23373
• [executorch][cuda] Let CudaExecutor honor a program's prefill lower bound #23372
• [executorch][llm] Share the batching executors' session logic under extension/llm/batching/util #23371
• [executorch][cuda] Add CudaExecutor #23370
• [executorch][cuda] Add step_plan for the CUDA batching executor #23369
• [executorch][llm] Add Buck targets for extension/llm/batching #23368
• [executorch][cuda] Add CudaCellCache #23367
• [executorch][cuda] Lower off-graph KV in a cell layout #23366
• [executorch][cuda] Extract CudaKVPool from CudaSequenceKVCache #23365

The runtime half of batched Muse Glimmer on CUDA: runs what
export_solo_batching.py writes, on the batching extension only. It is
separate from solo.cpp and MuseGlimmerEngine, which it leaves untouched. The
structure follows the MLX batched example (mlx_run_llm_batched), with
CudaExecutor in place of ModuleExecutor.

• One session per prompt. The batching Runner carries every generation, and
  DecodeFirstScheduler packs decodes and prefill chunks into shared forwards
  of up to W tokens (--max_decode_sequences decode slots, default one per
  session).
• Prompts come from --prompt_file (one verbatim, already formatted),
  --prompts_file (one per line), or --prompt_tokens_file (token ids, no
  tokenizer needed). Stop tokens are the model's EOS plus <|eot|> and
  <|end_of_text|>, as MuseGlimmerEngine uses.
• --cuda_graph and --weight_sharing (both default on) map to
  CudaExecutorOptions.
• Prints each prompt's text, engine counters, KV pool usage, and GPU memory
  before load, after load and after generation. --report_json writes all of
  it, with each generation's tokens and finish reason, for CI to check.

Builds as the run_solo_batching target of the MG CMake project and the
muse-glimmer-cuda preset/make target (CUDA only). It links cuda_batching and
exports dynamic symbols for the AOTI blobs, as solo_runner does.

The tiny-model e2e test joins the OSS unittest-cuda job, after
make muse-glimmer-cuda.

Differential Revision: D123051485

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/23374

• 📄 Preview Python docs built from this PR

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

❌ 6 New Failures, 1 Pending, 2 Unclassified Failures

As of commit 4677ca5 with merge base c16dd0a ():

NEW FAILURES - The following jobs have failed:

• pull / unittest / linux / linux-job (gh)
  backends/apple/coreml/test/test_torch_ops.py::TestTorchOps::test_dequantize_affine_c8w_embedding_c8w_linear_ios16
• pull / unittest / macos / macos-job (gh)
  backends/apple/coreml/test/test_torch_ops.py::TestTorchOps::test_dequantize_affine_c8w_embedding_c8w_linear_ios16
• pull / unittest / windows / windows-job (gh)
  .ci/scripts/tests/test_wheel_test_modules.py::TestDroppedTestModules::test_ci_entry_points_still_match_the_workflows
• pull / unittest-editable / linux / linux-job (gh)
  backends/apple/coreml/test/test_torch_ops.py::TestTorchOps::test_dequantize_affine_c8w_embedding_c8w_linear_ios16
• pull / unittest-editable / macos / macos-job (gh)
  backends/apple/coreml/test/test_torch_ops.py::TestTorchOps::test_dequantize_affine_c8w_embedding_c8w_linear_ios16
• pull / unittest-editable / windows / windows-job (gh)
  .ci/scripts/tests/test_wheel_test_modules.py::TestDroppedTestModules::test_ci_entry_points_still_match_the_workflows

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

• Test CUDA Builds / unittest-cuda / linux-job (gh) (this job did not run on the merge base, so DrCI cannot tell whether the failure is pre-existing)
  backends/cuda/tests/test_offgraph_kv.py::OffGraphKVCellCompileTest::test_decode_and_prefill_from_two_tokens_lower
• Test CUDA Builds / unittest-cuda-runtime / linux-job (gh) (this job did not run on the merge base, so DrCI cannot tell whether the failure is pre-existing)
  [ FAILED ] CudaKVCacheTest.CacheOutlivesTheStreamItLastSteppedOn

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.