Remove pad_max_tiles from preprocess

[lucylq]

Following changes in torchtune:

• Remove pad_max_tiles in CLIP meta-pytorch/torchtune#1836
• Remove pad_max_tiles in CLIP inference meta-pytorch/torchtune#1853

Update ET downstream and remove pad-max-tiles from preprocess.

Test Plan:
With AOTI tests commented out (not working atm), so testing eager/export/et:

python -m unittest examples/models/llama3_2_vision/preprocess/test_preprocess.py 
...

----------------------------------------------------------------------
Ran 4 tests in 21.129s

OK

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/6295

• 📄 Preview Python docs built from this PR

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

• [PREEMTIVE] Experimenting with new runners linux.aws.a100 on inductor-perf-compare.yml

✅ No Failures

As of commit 0bbd026 with merge base 5c3439d ():
💚 Looks good so far! There are no failures yet. 💚

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

[facebook-github-bot]

@lucylq has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

[facebook-github-bot]

@lucylq merged this pull request in 2e67e3a.