Disable quick init for TapasPreTrainedModel

[daniellok-db]

What does this PR do?

Fixes #32128

There are more details in the linked issue, but it seems that the superfast init feature causes this pipeline to fail saving with the following error:

ValueError: You are trying to save a non contiguous tensor: 
`tapas.encoder.layer.0.attention.self.query.weight` which is not allowed. 
It either means you are trying to save tensors which are reference of each
 other in which case it's recommended to save only the full tensors, and 
reslice at load time, or simply call `.contiguous()` on your tensor to pack 
it before saving.

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline,
  Pull Request section?
• Was this discussed/approved via a Github issue or the forum? Please add a link
  to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
• Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

@amyeroberts @muellerzr

[muellerzr]

Thanks!

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

[amyeroberts]

Thanks for fixing this!