[Bug Fix] Wan2.1 runnable with MS2.6/2.7

[zhtmike]

What does this PR do?

Wan 2.1 Runnable with MindSpore 2.6/2.7
Wan 2.2 Runnable with MindSpore 2.7

Fixes # (issue)

1. Fix the missing scalar value in clip and xlm_roberta, causing the output slightly inconsistent.
2. Fix the output of the prompt extender due to the recent update of mindone.transformers.
3. Change attention to FA for Wan2.1/Wan2.2 for prompt extender for memory saving.

Adds # (feature)

1. Update the README for MS2.6/2.7

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?
• Did you make sure to update the documentation with your changes? E.g. record bug fixes or new features in What's New. Here are the
  documentation guidelines
• Did you build and run the code without any errors?
• Did you report the running environment (NPU type/MS version) and performance in the doc? (better record it for data loading, model inference, or training tasks)
• 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.

@xxx

[gemini-code-assist]

Summary of Changes

Hello @zhtmike, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request focuses on enhancing the stability and efficiency of the wan2_1 and wan2_2 examples. It resolves critical bugs related to attention mechanisms and prompt extender output, ensuring more consistent and accurate results. Additionally, it introduces performance optimizations by adopting Flash Attention for the prompt extender models, which helps in reducing memory consumption.

Highlights

• Bug Fix: Missing Scalar Value in Attention: Corrected an issue in clip.py and xlm_roberta.py by adding a scalar_value to flash_attention_score operations, resolving output inconsistencies that could arise from its absence.
• Bug Fix: Prompt Extender Output Inconsistency: Addressed an issue in wan2_1/wan/utils/prompt_extend.py related to recent mindone.transformers updates, ensuring correct trimming of generated output and proper tensor conversion.
• Performance Improvement: Flash Attention for Prompt Extender: Switched the attention implementation to flash_attention_2 for Qwen2ForCausalLM and AutoModelForCausalLM in both wan2_1 and wan2_2 prompt extenders, leading to significant memory savings and improved efficiency.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request introduces several bug fixes and performance improvements across the wan examples. Key changes include adding the necessary scaling factor to flash_attention_score calls, which is a critical correctness fix for the attention mechanism. The prompt extender logic is also updated to fix tensor conversion and output trimming issues. Additionally, switching to flash_attention_2 is a welcome memory optimization. The changes are well-aligned with the PR's description. I've added a few suggestions to further improve performance and code clarity.