Qualcomm AI Engine Direct - issue fix #2

[pytorchbot]

Summary

• Conformer model fails to lower on QNN #14048 > add quantized test case with GLU decomposition
• Convnext model fails to lower on QNN #14049 > add e2e example where constant expansion is applied
• Maxvit_t model lowering segfaults on QNN #14050 > add e2e example and source transform for 6D operation
• Vit_b_16 model fails to lower on QNN #14051 > add e2e example and complement missed annotation
• Swin_v2_t model fails to lower on QNN #14052 > add e2e example and dedicated passe for 6D partition

Fixes #14048
Fixes #14049
Fixes #14050
Fixes #14051
Fixes #14052

Test plan

MATRIX = {convnext_small, maxvit_t, swin_v2_t, vit_b_16}

python backends/qualcomm/tests/test_qnn_delegate.py TestExampleOssScript.test_${MATRIX} -b build-android/ -m SM8750 -s $SN -a /path/to/test_artifacts/ -i /path/to/imagenet_1k/imagenet-mini/val -r .

python backends/qualcomm/tests/test_qnn_delegate.py TestQuantizedModel.test_qnn_backend_conformer -b build-android/ -m SM8750 -s $SN -a /path/to/test_artifacts/

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/14947

• 📄 Preview Python docs built from this PR

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

❌ 2 Cancelled Jobs

As of commit 58c69ee with merge base e0dda90 ():

CANCELLED JOBS - The following jobs were cancelled. Please retry:

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

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