[executorch][cuda] Let CudaExecutor honor a program's prefill lower bound

[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

Muse Glimmer's quantized linears switch kernels at M <= 4 (the int4/5/6/8
dispatch in backends/cuda/quantize_op_dispatch). A prefill with a symbolic
token count therefore only traces for T >= 5, and the selected-logits rows
feeding the LM head for K >= 5. This is why the single-sequence MG export
starts its dynamic method at 5.

CudaExecutor now reads the optional get_min_prefill_chunk constant (the name
the MG export already uses; default 2):

• plan_slices takes it as the prefill lower bound, so slices narrower than it
  run as that many decode forwards;
• a prefill slice selecting fewer rows than the bound pads its selector by
  repeating the last row, and the padded rows are ignored.

create() refuses a bound outside [2, W].

Differential Revision: D123051483

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/23372

• 📄 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, 3 Unclassified Failures

As of commit a0a7329 with merge base c16dd0a ():

NEW FAILURES - The following jobs have failed:

• pull / unittest / linux / linux-job (gh)
  .ci/scripts/tests/test_wheel_test_modules.py::TestDroppedTestModules::test_ci_entry_points_still_match_the_workflows
• pull / unittest / macos / macos-job (gh)
  .ci/scripts/tests/test_wheel_test_modules.py::TestDroppedTestModules::test_ci_entry_points_still_match_the_workflows
• 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)
  .ci/scripts/tests/test_wheel_test_modules.py::TestDroppedTestModules::test_ci_entry_points_still_match_the_workflows
• pull / unittest-editable / macos / macos-job (gh)
  .ci/scripts/tests/test_wheel_test_modules.py::TestDroppedTestModules::test_ci_entry_points_still_match_the_workflows
• 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
• Test ROCm AOTI / unittest-rocm-gfx950-rocm7.2 / 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

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.