Support multiple batches (N > 1) in Arm runtime

[AdrianLundell]

Extends testing of multiple batches on FVP for multiple operators. Note that currently not all operators are supported on compiler level.

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/7358

• 📄 Preview Python docs built from this PR

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

❌ 3 New Failures

As of commit f481b12 with merge base eca5d9f ():

NEW FAILURES - The following jobs have failed:

• pull / unittest / macos / macos-job (gh)
  examples/models/llama3_2_vision/vision_encoder/test/test_vision_encoder.py::FlamingoVisionEncoderTest::test_flamingo_vision_encoder
• pull / unittest-arm / linux-job (gh)
  ../../dev/ops/test_softmax.py::TestSoftmax::test_softmax_tosa_BI_7_randn_neg_dim
• trunk / test-arm-reference-delegation / linux-job (gh)
  ../../dev/ops/test_softmax.py::TestSoftmax::test_softmax_tosa_BI_7_randn_neg_dim

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

[AdrianLundell]

The failing tests should be marked as flaky since they are failing randomly without any related changes, looking into it.

[freddan80]

Rebase pls

[AdrianLundell]

We'll wait with this PR until the softmax failures are better understood

[AdrianLundell]

Closing this PR as we are likely to move transposes into the delegate anyways.