ofa增加asr任务infer
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10761019

为diffusion模型加入dpm solver支持，相比ddim scheduler快2～6倍。
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10826722

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10851834

    * [to #46342279]fix: fix some time test_skin_retouching.py failed

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10814720

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10773667

It is recommented that each pipleine should implement `_batch`  to make a list of preprocessed data into a batched data dict.

Then  by paasing batch_size=n  we can use batch inference in pipline, for example
```python
img_captioning = pipeline(
            Tasks.image_captioning,
            model='damo/ofa_image-caption_coco_large_en')

results = img_captioning(
            [{
                'image': 'data/test/images/image_captioning.png'
            } for _ in range(6)],
            batch_size=2)
```

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10051193

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10834768

新增对ofa多GPU训练的支持
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10838906

1. add damoyolo-t & damoyolo-m models
2. fix the configuration overlap error
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10816561

Merge master internal 1118

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10780767

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10778739

    * [to #46219277]fix pre-commit pycqa flake8 to github

fix 输出为空的情况
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10725840

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10763022

    * [fix sequence labeling postprocess bug]

Merge master internal  to master

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10750369

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10743508

* fix: load model directly from .pth

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10636269

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10728515

    * make mglm isolated test because mpu init problem

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10713843

    * [to #42322933] solve memory error for translation finetune

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10708723

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10712244

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10708368

提高ci效率，先跑linter，linter错误先报漏出来，不用合并release
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10596907

    * [to #45878116]feat: run linter test first

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10617428

在modelscope 基础镜像中添加 jupyter lab 插件，用于收集用户在notebook 中的鼠标、键盘事件，来支持静默关闭功能。
主要改变如下：
1. 新增jupyterlab_active_log 插件的源码和编译后的文件
2. 修改Dockerfile.ubuntu 文件，将上述源码文件拷贝的jupyter的插件目录中。
3. 修改pip和conda 的镜像地址为阿里云的
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10690152

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10680402

主要做了如下修改：
1. 加入了同义词词典
2. 对SQL进行后处理，如果包含排序，则将空列转化成Primary列
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10670121