Fix the Windows Voxtral Realtime CI test (#24289)

The Windows Voxtral Realtime test could never pass, for two reasons.
Both came in with #17993, which added `-Mode` to
`test_model_e2e_windows.ps1` with streaming as the default and switched
the expected word to "Quilter", without updating the workflow that calls
the script.

- The workflow exports the model in offline mode but called the test
without `-Mode`, so the script ran the runner with `--streaming`. The
runner refuses `--streaming` for a model exported without it and aborts.
Release builds compile out the log, so the job only showed exit code
0xC0000409. The mode now lives in one place: a `mode` key on the Voxtral
Realtime row of both matrices, passed by the export and the test step
alike. Other rows have no key and pass an empty mode, which both scripts
treat as before.
- The test looked for "Quilter" in a transcript of a poem recording that
does not contain that word. "Quilter" comes from the LibriSpeech clip
the Linux test uses. The test now uses the LibriSpeech clip the Windows
Parakeet test already downloads (saved as `test_audio.wav` instead of
`poem.wav`) and looks for "Phoebe", as the MLX Voxtral Realtime test
does with the same clip.

A second commit adds `export_model_artifact.sh` and
`test_model_e2e_windows.ps1` to the workflow's paths filter and to the
export and test job conditions, as cuda.yml does for its own scripts.
Before, a change that only touched those scripts ran no Windows model
job.

Testing: on Linux CUDA, a Release runner with logging off reproduces the
Windows abort with the old flags, and with the new ones transcribes the
clip and prints "Phoebe". This branch lives in the main repository so
the Windows CUDA jobs run on this PR; on the previous head all six
Windows model tests passed, including Voxtral Realtime with "Phoebe".

Co-authored-by: Anthony Shoumikhin <anthony@shoumikh.in>

Author: pytorchbot

.ci/scripts/test_model_e2e_windows.ps1

@@ -89,12 +89,12 @@ switch ($HfModel) {
         $runnerTarget = "voxtral_realtime_runner"
         $runnerPath = "voxtral_realtime"
         $runnerPreset = "voxtral-realtime-cuda"
-        $expectedOutput = "Quilter"
+        $expectedOutput = "Phoebe"
         $preprocessor = "preprocessor.pte"
         $tokenizerUrl = ""
         $tokenizerFile = "tekken.json"
-        $audioUrl = "https://github.com/voxserv/audio_quality_testing_samples/raw/refs/heads/master/testaudio/16000/test01_20s.wav"
-        $audioFile = "poem.wav"
+        $audioUrl = "https://dldata-public.s3.us-east-2.amazonaws.com/2086-149220-0033.wav"
+        $audioFile = "test_audio.wav"
     }
     "facebook/dinov2-small-imagenet1k-1-layer" {
         $runnerTarget = "dinov2_runner"

.github/workflows/cuda-windows.yml

@@ -17,6 +17,8 @@ on:
       - backends/cuda/**
       - backends/aoti/**
       - extension/cuda/**
+      - .ci/scripts/export_model_artifact.sh
+      - .ci/scripts/test_model_e2e_windows.ps1
       # Only unittest-cuda-runtime-windows runs for these alone; the other jobs
       # filter on their own paths and stay skipped.
       - .ci/scripts/test_cuda_runtime_windows.ps1
@@ -61,6 +63,8 @@ jobs:
         contains(needs.changed-files.outputs.changed-files, 'backends/aoti') ||
         contains(needs.changed-files.outputs.changed-files, 'extension/cuda') ||
         contains(needs.changed-files.outputs.changed-files, '.github/workflows/cuda-windows.yml') ||
+        contains(needs.changed-files.outputs.changed-files, '.ci/scripts/export_model_artifact.sh') ||
+        contains(needs.changed-files.outputs.changed-files, '.ci/scripts/test_model_e2e_windows.ps1') ||
         needs.run-decision.outputs.is-full-run == 'true'
       )
     uses: pytorch/test-infra/.github/workflows/linux_job_v3.yml@main
@@ -92,6 +96,7 @@ jobs:
           - model_repo: "mistralai"
             model_name: "Voxtral-Mini-4B-Realtime-2602"
             quant: "quantized-int4-tile-packed"
+            mode: "vr-offline"
           - model_repo: "facebook"
             model_name: "dinov2-small-imagenet1k-1-layer"
             quant: "non-quantized"
@@ -147,11 +152,7 @@ jobs:
           echo "::endgroup::"
         fi
 
-        VR_MODE=""
-        if [ "${{ matrix.model_name }}" = "Voxtral-Mini-4B-Realtime-2602" ]; then
-          VR_MODE="vr-offline"
-        fi
-        source .ci/scripts/export_model_artifact.sh cuda-windows "${{ matrix.model_repo }}/${{ matrix.model_name }}" "${{ matrix.quant }}" "${RUNNER_ARTIFACT_DIR}" "${VR_MODE}"
+        source .ci/scripts/export_model_artifact.sh cuda-windows "${{ matrix.model_repo }}/${{ matrix.model_name }}" "${{ matrix.quant }}" "${RUNNER_ARTIFACT_DIR}" "${{ matrix.mode }}"
 
   test-model-cuda-windows-e2e:
     name: test-model-cuda-windows-e2e
@@ -168,6 +169,8 @@ jobs:
         contains(needs.changed-files.outputs.changed-files, 'backends/aoti') ||
         contains(needs.changed-files.outputs.changed-files, 'extension/cuda') ||
         contains(needs.changed-files.outputs.changed-files, '.github/workflows/cuda-windows.yml') ||
+        contains(needs.changed-files.outputs.changed-files, '.ci/scripts/export_model_artifact.sh') ||
+        contains(needs.changed-files.outputs.changed-files, '.ci/scripts/test_model_e2e_windows.ps1') ||
         needs.run-decision.outputs.is-full-run == 'true'
       )
     uses: pytorch/test-infra/.github/workflows/windows_job.yml@main
@@ -195,6 +198,7 @@ jobs:
           - model_repo: "mistralai"
             model_name: "Voxtral-Mini-4B-Realtime-2602"
             quant: "quantized-int4-tile-packed"
+            mode: "vr-offline"
           - model_repo: "facebook"
             model_name: "dinov2-small-imagenet1k-1-layer"
             quant: "non-quantized"
@@ -228,7 +232,7 @@ jobs:
             throw 'RUNNER_ARTIFACT_DIR is empty. Ensure download-artifact is configured for windows_job.yml.'
           }
 
-          .ci/scripts/test_model_e2e_windows.ps1 -Device cuda-windows -HfModel '${{ matrix.model_repo }}/${{ matrix.model_name }}' -QuantName '${{ matrix.quant }}' -ModelDir \$artifactDir -ExpectedCudaVersion '13.2'
+          .ci/scripts/test_model_e2e_windows.ps1 -Device cuda-windows -HfModel '${{ matrix.model_repo }}/${{ matrix.model_name }}' -QuantName '${{ matrix.quant }}' -ModelDir \$artifactDir -ExpectedCudaVersion '13.2' -Mode '${{ matrix.mode }}'
         }"
 
   # The counterpart of cuda.yml's unittest-cuda-runtime on Windows. The Windows GPU runners