[Parakeet] Add Vulkan backend documentation and fix CMake build

[SS-JIA]

Summary:
Add Vulkan backend documentation to the Parakeet README covering export
commands, quantization options, build instructions, and runner examples.

Guard quantized_ops_lib and custom_ops link targets with
if(TARGET ...) in CMakeLists.txt. These targets don't exist in
Vulkan-only or XNNPACK-only builds, causing a hard CMake configure
error from target_link_options(). This matches the existing pattern
used for optimized_native_cpu_ops_lib.

Validated on Samsung S24 (Adreno 750), 8da4w quantization, test_audio.wav (7.2s):

Metric	XNNPACK (686 MB)	Vulkan (781 MB)	Vulkan fp16 (550 MB)
Inference	0.56s	0.46s	0.32s
Encoder speed	188 tok/s	275 tok/s	360 tok/s
Decoder speed	657 tok/s	373 tok/s	746 tok/s

Authored by Claude (Anthropic)

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/18734

• 📄 Preview Python docs built from this PR

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

❌ 1 New Failure, 3 Unrelated Failures

As of commit df72cc1 with merge base fcccda3 ():

NEW FAILURE - The following job has failed:

• Cadence Build & Test / cpu-test / test-ops / test-ops (gh)
  examples/cadence/operators/test_g3_ops.py::ATenOpTestCases::test_g3_neg_out_7

FLAKY - The following job failed but was likely due to flakiness present on trunk:

• pull / test-models-linux-basic (mv3, xnnpack-quantization-delegation, cmake, linux.2xlarge, executorch-u... / linux-job (gh) (detected as infra flaky with no log or failing log classifier)

BROKEN TRUNK - The following jobs failed but was present on the merge base:

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

• pull / unittest / windows / windows-job (gh) (trunk failure)
  ##[error]The operation was canceled.
• pull / unittest-editable / windows / windows-job (gh) (trunk failure)
  ##[error]The operation was canceled.

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.

[mergennachin]

We cant use --dtype flag?

[SS-JIA]

• --dtype fp16: inputs and outputs are also cast to fp16. From caller's perspective, input/output is fp16.
• --vulkan_force_fp16: inputs and outputs are still fp32. Vulkan backend will automatically convert inputs to fp16 within the delegate and outputs to fp32. From caller's perspective, input/output is fp32.

--vulkan_force_fp16 is a bit simpler for client code since they don't have to handle the conversion to/from fp32, so I defaulted to that.

Another thing was that for the export_parakeet_tdt.py script, with --dtype fp16 I see there is a guard:

export_parakeet_tdt.py: error: fp16 is not yet supported

I wasn't sure if this was because the runner binary doesn't handle fp16 input/output yet, so I opted for --vulkan_force_fp16 instead.

Would you prefer enabling usage of the --dtype flag for fp16 inference?

[SS-JIA]

Also updated the text to clarify the special properties of the --vulkan_force_fp16 flag that wouldn't be covered by --dtype