[fix] [whisper] ensure inputs are moved to the correct device before processing.

[AgainstEntropy]

Motivation

Whisper was not covered by the Lazy device transfer introduced by #22038 , thus caused a bug that the input features may not be transferred to the correct device.
Can be spotted by the manual test here: https://github.com/sgl-project/sglang/blob/main/test/manual/test_whisper_cuda_graph.py

python test/manual/test_whisper_cuda_graph.py

• main branch

  ...
  inputs_embeds = torch.nn.functional.gelu(self.conv1(input_features))
  ...
  return F.conv1d(
  ...
  RuntimeError: Expected all tensors to be on the same device, but got weight is on cuda:0, different from other tensors on cpu (when checking argument in method wrapper_CUDA___slow_conv2d_forward)

• this PR
  everything good

Modifications

Accuracy Tests

Speed Tests and Profiling

Checklist

• Format your code according to the Format code with pre-commit.
• Add unit tests according to the Run and add unit tests.
• Update documentation according to Write documentations.
• Provide accuracy and speed benchmark results according to Test the accuracy and Benchmark the speed.
• Follow the SGLang code style guidance.

Review and Merge Process

1. Ping Merge Oncalls to start the process. See the PR Merge Process.
2. Get approvals from CODEOWNERS and other reviewers.
3. Trigger CI tests with comments or contact authorized users to do so.
   • Common commands include /tag-and-rerun-ci, /tag-run-ci-label, /rerun-failed-ci
4. After green CI and required approvals, ask Merge Oncalls or people with Write permission to merge the PR.

[gemini-code-assist]

Warning

You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again!

[yhyang201]

/tag-and-rerun-ci

[mickqian]

nit: device=next(self()).device is more robust

[yhyang201]

all cuda ci passed.