Replace AVG_POOL2D with REDUCE_SUM in DecomposeMeanDimPass

[mcremon-meta]

Summary:
Replace the avg_pool2d decomposition path in DecomposeMeanDimPass with
REDUCE_SUM + MUL(1/N) for all mean.dim reductions.

AVG_POOL2D can only pool over spatial (H×W) axes in TOSA/NHWC layout,
which forces the compiler to insert TRANSPOSE ops when the reduction is
over channels (common in LayerNorm). REDUCE_SUM works on any axis without
layout constraints, avoiding those transposes entirely.

Reviewed By: 3l1

Differential Revision: D101418199

cc @digantdesai @freddan80 @per @zingo @oscarandersson8218 @mansnils @Sebastian-Larsson @robell

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/19242

• 📄 Preview Python docs built from this PR

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

❌ 5 New Failures, 3 Unrelated Failures

As of commit 8dff93f with merge base 94d2881 ():

NEW FAILURES - The following jobs have failed:

• pull / unittest-arm-backend-with-no-deps (test_pytest_ops_tosa) / linux-job (gh)
  RuntimeError: Command docker exec -t cc80f27d069543f4e81a43bcffbbfd1d89adfdfe1a55f7a4c9e75c0bbfb18420 /exec failed with exit code 1
• trunk / test-arm-backend-ethos-u (test_pytest_ops_ethos_u85) / linux-job (gh)
  RuntimeError: Command docker exec -t 9f9970c0ed0c5cf53f49bf31d2b194a40f72006057642de68702f8758683a925 /exec failed with exit code 1
• trunk / test-llama-runner-mac (fp32, xnnpack+custom+quantize_kv) / macos-job (gh)
  RuntimeError: Command bash /Users/ec2-user/runner/_work/_temp/exec_script failed with exit code 1
• trunk / test-models-macos-coreml (resnet50) / macos-job (gh)
  RuntimeError: Command bash /Users/ec2-user/runner/_work/_temp/exec_script failed with exit code 1
• trunk / test-models-macos-cpu (w2l, xnnpack-quantization-delegation) / macos-job (gh)
  RuntimeError: Command bash /Users/ec2-user/runner/_work/_temp/exec_script failed with exit code 1

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

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

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

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

[meta-codesync]

@mcremon-meta has exported this pull request. If you are a Meta employee, you can view the originating Diff in D101418199.

[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.

[gggekov]

Thanks for the PR, Matthias! I see a few failing tests, could you take a look into them?
You are indeed reducing the number of permutes in a few cases, can you please update the lower number of expected permutes in backends/arm/test/misc/test_transpose_counts.py ?
I am also seeing a failure in backends/arm/test/ops/test_layer_norm.py for the test_native_layer_norm_16a8w_u85_INT suite and for backends/arm/test/ops/test_cond -k test_cond_u85_INT for the one_arg_two_outputs test case, wonder why is that?