[LLM INFER] Append attn

[yuanlehome]

PR types

New features

PR changes

Others

Description

大模型推理attention组网重构，新的append_attn方案相比旧方案有10%到90%的性能提升。

目前已支持了llama/qwen/qwen-moe/mixtral的推理。

使用方式，原推理脚本的 --block_attn选项改为--append_attn即可。

TODO：

• fp8推理适配
• 性能数据补充，稍后见llm docs

[paddle-bot]

Thanks for your contribution!

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 60 lines in your changes missing coverage. Please review.

Project coverage is 53.08%. Comparing base (04f3c20) to head (4011d89).
Report is 13 commits behind head on develop.

Files with missing lines	Patch %	Lines
...erimental/transformers/fused_transformer_layers.py	0.00%	38 Missing ⚠️
...dlenlp/experimental/transformers/qwen2/modeling.py	0.00%	8 Missing ⚠️
...dlenlp/experimental/transformers/llama/modeling.py	0.00%	7 Missing ⚠️
...enlp/experimental/transformers/mixtral/modeling.py	0.00%	5 Missing ⚠️
...lp/experimental/transformers/qwen2_moe/modeling.py	0.00%	1 Missing ⚠️
paddlenlp/experimental/transformers/utils.py	0.00%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #9244      +/-   ##
===========================================
+ Coverage    52.81%   53.08%   +0.26%     
===========================================
  Files          660      657       -3     
  Lines       107281   106801     -480     
===========================================
+ Hits         56660    56690      +30     
+ Misses       50621    50111     -510     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[vivienfanghuagood]

中文注释删一删

[yuanlehome]

太多了，留着无伤大雅吧