Add CUDA-Windows quantization support to Parakeet README

[seyeong-han]

Document that the cuda-windows backend supports 4w/8w/8da4w/8da8w quantization configs for the Parakeet TDT model export, and add an export example.

Changes

• Updated the Quantization Configs table to list CUDA-Windows alongside CUDA as a supported backend for 4w, 8w, 8da4w, and 8da8w quantization
• Added a new "Example: 4-bit Weight Quantization for CUDA-Windows" section with the export command (--dtype bf16 --qlinear_encoder 4w --qlinear 4w --qembedding 8w)
• Added a note that --qlinear_packing_format tile_packed_to_4d is not supported for cuda-windows because the cross-compilation path keeps tensors on CPU during export and aten::_convert_weight_to_int4pack only has a CUDA kernel

Benchmark Results (RTX 5080, ~20s audio)

Metric	Non-Quantized (fp32)	Quantized (bf16 + 4w/8w)	Improvement
Model size	2,445 MB	763 MB	3.2× smaller
Prefill speed	1,301 tok/s	3,218 tok/s	2.5× faster
Decode speed	1,325 tok/s	1,545 tok/s	1.2× faster
Model load time	4.1 s	1.6 s	2.5× faster
Time to first token	202 ms	86 ms	2.4× faster
Transcription accuracy	Baseline	Identical words	Tie

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/18895

• 📄 Preview Python docs built from this PR

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

• Rolling out OSDC (ARC) runners on pull workflow for PyTorch trunk commits

✅ No Failures

As of commit 7f25285 with merge base 38c5ca3 ():
💚 Looks good so far! There are no failures yet. 💚

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.