[LLM] support QWen1.5-Moe

[DrownFish19]

PR types

New features

PR changes

Models

Description

1. add QWen1.5 Moe model.
2. support same prefix for different models, such as QWen and QWen2Moe with same prefix QWen. The longest name will match each model name before others.
3. support sft and lora.

[paddle-bot]

Thanks for your contribution!

[codecov]

Codecov Report

Attention: Patch coverage is 66.62959% with 300 lines in your changes are missing coverage. Please review.

Project coverage is 55.47%. Comparing base (3aa92ce) to head (2bb3aba).
Report is 18 commits behind head on develop.

❗ Current head 2bb3aba differs from pull request most recent head 58af3ec. Consider uploading reports for the commit 58af3ec to get more accurate results

Files	Patch %	Lines
paddlenlp/transformers/qwen2moe/modeling.py	72.33%	197 Missing ⚠️
paddlenlp/transformers/qwen2moe/tokenizer.py	23.13%	103 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #8338      +/-   ##
===========================================
+ Coverage    55.36%   55.47%   +0.10%     
===========================================
  Files          614      618       +4     
  Lines        96016    96913     +897     
===========================================
+ Hits         53164    53761     +597     
- Misses       42852    43152     +300     

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