Fix warning raised by torch.meshgrid on missing indexing

[vfdev-5]

Description:

Calling torch.meshgrid(x, y) without indexing argument raises this warning:

[/opt/conda/lib/python3.11/site-packages/torch/functional.py:513](http://localhost:2211/lab/tree/jax-ai-stack/docs/opt/conda/lib/python3.11/site-packages/torch/functional.py#line=512): UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1720538435607/work/aten/src/ATen/native/TensorShape.cpp:3609.)
  return _VF.meshgrid(tensors, **kwargs)  # type: ignore[attr-defined]

This PR added indexing with its default value in the places where it is missing.

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/vision/8689

• 📄 Preview Python docs built from this PR

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

❌ 2 New Failures, 1 Unrelated Failure

As of commit c42d5f8 with merge base 945bdad ():

NEW FAILURES - The following jobs have failed:

• Lint / c-source / linux-job (gh)
  RuntimeError: Command docker exec -t bf59126c9a7deb6279052d47559a95485f397e7a4db90839f3ef3d5da7a497e7 /exec failed with exit code 1
• Tests / unittests-windows (3.11, windows.4xlarge, cpu) / windows-job (gh)
  test/test_datasets.py::FlyingThings3DTestCase::test_num_examples

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

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

• Tests / unittests-windows (3.9, windows.g5.4xlarge.nvidia.gpu, cuda, 11.8) / windows-job (gh) (trunk failure)
  AttributeError: '_OpNamespace' 'image' object has no attribute '_jpeg_version'

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