Integrate CUDA sampler into ASR runner and enable skip_copy for decoder

[larryliu0820]

• cuda_backend.cpp: Support comma-separated method names in
  skip_copy_output_to_cpu_for_method backend option.
• runner.cpp: Use CudaSampler for argmax when CUDA is available and
  temperature==0. Skip copy to CPU for both encoder and decoder methods.
• CMakeLists.txt updates: Link against extension_llm_sampler_cuda library
  and include the sampler subdirectory.

This optimization keeps decoder logits on GPU and performs argmax directly
on GPU memory, avoiding unnecessary device-to-host copies in the decode loop.

[larryliu0820]

Stack from ghstack (oldest at bottom):

• -> Integrate CUDA sampler into ASR runner and enable skip_copy for decoder #16388
• Add CudaSampler class for GPU-based token sampling #16387
• Add CUDA argmax kernel for LLM sampler #16386

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/16388

• 📄 Preview Python docs built from this PR

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

❌ 14 New Failures, 2 Unrelated Failures

As of commit 1637ec1 with merge base c5d66a5 ():

NEW FAILURES - The following jobs have failed:

• Lint / lintrunner / linux-job (gh)
  >>> Lint for extension/llm/runner/CMakeLists.txt:
• pull / test-multimodal-linux (gemma3-4b) / linux-job (gh)
  RuntimeError: Command docker exec -t 4466472f45a0b195db5e372715a237cc95d589061175d6bb474a95673f9bbac0 /exec failed with exit code 139
• Test CUDA Builds / test-cuda-shims / linux-job (gh)
  RuntimeError: Command docker exec -t ac2acada201032fcfa28a448f30239f80f8d56cd6f444e4b6f1fa8954029d86a /exec failed with exit code 1
• Test CUDA Builds / test-model-cuda-e2e (google, gemma-3-4b-it, non-quantized) / linux-job (gh)
  RuntimeError: Command docker exec -t 42cc3d4baf40e21817fc34e2c4b7de2c28d76dcdbef427a9d5066df17a7d7683 /exec failed with exit code 2
• Test CUDA Builds / test-model-cuda-e2e (google, gemma-3-4b-it, quantized-int4-tile-packed) / linux-job (gh)
  RuntimeError: Command docker exec -t 98ce809db8f343dce7bfe1f3435d7a3fd78d61988073ddfb2ccd28f56adf7224 /exec failed with exit code 2
• Test CUDA Builds / test-model-cuda-e2e (mistralai, Voxtral-Mini-3B-2507, non-quantized) / linux-job (gh)
  RuntimeError: Command docker exec -t 159d6f51c05833debfef29c964c94eeb9872bdddb6867993b71ca52171868ad1 /exec failed with exit code 2
• Test CUDA Builds / test-model-cuda-e2e (mistralai, Voxtral-Mini-3B-2507, quantized-int4-tile-packed) / linux-job (gh)
  RuntimeError: Command docker exec -t dcfa35d3dff75b5e3f1516adf19577b960b2324a59e0ef526b2f078519b4d97c /exec failed with exit code 2
• Test CUDA Builds / test-model-cuda-e2e (mistralai, Voxtral-Mini-3B-2507, quantized-int4-weight-only) / linux-job (gh)
  RuntimeError: Command docker exec -t a5bb4a490b58993bed9eb0b97b087ef390fb8b8391528e4ac0abb7c718b6bdac /exec failed with exit code 2
• Test CUDA Builds / test-model-cuda-e2e (openai, whisper-large-v3-turbo, non-quantized) / linux-job (gh)
  RuntimeError: Command docker exec -t 51541149242cbad68000b1caf53b1327838776ba66a66650d7f388f842775a4c /exec failed with exit code 2
• Test CUDA Builds / test-model-cuda-e2e (openai, whisper-large-v3-turbo, quantized-int4-tile-packed) / linux-job (gh)
  RuntimeError: Command docker exec -t 5815100b569dbe333addcd54e15cf3dbf583ffd277de780fae5e0698b777dd74 /exec failed with exit code 2
• Test CUDA Builds / test-model-cuda-e2e (openai, whisper-large-v3-turbo, quantized-int4-weight-only) / linux-job (gh)
  RuntimeError: Command docker exec -t 126e9de5984bcb7991384a27156e517f945f9ab362f5373e408939e37d0458bd /exec failed with exit code 2
• Test CUDA Builds / test-model-cuda-e2e (openai, whisper-small, non-quantized) / linux-job (gh)
  RuntimeError: Command docker exec -t 8d165dbedd1e24b5fe4567656d62d904dd6dd3cec3b486bc0f15681aebf66798 /exec failed with exit code 2
• Test CUDA Builds / test-model-cuda-e2e (openai, whisper-small, quantized-int4-tile-packed) / linux-job (gh)
  RuntimeError: Command docker exec -t be9f5859fd05da90e00768e8dc94de3afb693a0af847346062ca11cb5ccacca2 /exec failed with exit code 2
• Test CUDA Builds / test-model-cuda-e2e (openai, whisper-small, quantized-int4-weight-only) / linux-job (gh)
  RuntimeError: Command docker exec -t 81e642c9af3e56071f8c2ead047550a96b163d4da2c750b86f7dcb2dcf63df81 /exec failed with exit code 2

BROKEN TRUNK - The following job failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

• pull / test-samsung-models-linux / linux-job (gh) (trunk failure)
  ##[error]The operation was canceled.

UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:

• pull / android / run-emulator (gh) (#16137)
  Timeout waiting for emulator to boot.

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

[Gasoonjia]

hmm can we make the skip_copy_method_ as an array? Maybe we can avoid the fancy comparsion here for supporting multiple method names