[bugfix] fix flash_attention_2 unavailable error on Ascend NPU

[FightingZhen]

What does this PR do?

After PR #39474 merged, flash_attention_2 is unavailable on Ascend NPU, because package flash-attn can not be installed on Ascend NPU :(

This PR is committed for solving this problem.

Detail modification:

1. Follow newest flash attention preparation logics in src/transformers/modeling_flash_attention_utils.py module, use get_npu_flash_attn_funcs return flash attention related functions for Ascend NPU.
2. Ignore logics related with package flash-attn in func _flash_attn_2_can_dispatch from src/transformers/modeling_utils.py module, it is not required for Ascend NPU.

Fixes # (issue)
Not related.

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?

[FightingZhen]

Please help me review this, thanks a lot! @ArthurZucker

[FightingZhen]

@ArthurZucker please help me review and merge, thanks ~

[Rocketknight1]

cc @ivarflakstad @SunMarc

[SunMarc]

Thanks ! Sorry for the breaking change

[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.

[Yulv-git]

Very good work. Can Flash Attention 3 also be used on the Ascend NPU?

[FightingZhen]

Very good work. Can Flash Attention 3 also be used on the Ascend NPU?

Hi, as far as i know, flash-attention-3 is not supported yet on Ascend NPU, see doc