[PaddleNLP 3.0] Refactor RLHF

[gongel]

PR types

Refactor codes

PR changes

Others

Description

• Refactor RLHF: mv examples/RLHF to llm/Alignment

[paddle-bot[bot]]

Thanks for your contribution!

[codecov[bot]]

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 54.42%. Comparing base (c134a7f) to head (281075d). Report is 3 commits behind head on develop.

Additional details and impacted files

```diff @@ Coverage Diff @@ ## develop #8617 +/- ## ======================================== Coverage 54.42% 54.42% ======================================== Files 632 632 Lines 99481 99481 ======================================== Hits 54140 54140 Misses 45341 45341 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[ZHUI]

下个pr加入 data 相关脚本。

[gongel]

下个pr加入 data 相关脚本。

data加过了，这里的gitignore需要后续删除。