Make TTAModel compatible with FSDP.

[HAOCHENYE]

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.

Motivation

1. Fix BaseTTAModel cannot set the device for BaseModel.
2. Provide a workaround to use FSDP and TTA simultaneously.
3. Add English and Chinese tutorial for TTA.
4. Make DataPreprocessor could accept non-tensor and non-datasample objects.

Related PR:

open-mmlab/mmdetection#9452
open-mmlab/mmsegmentation#2335
open-mmlab/mmpretrain#1161

Modification

1. Make BaseTTAModel inherit from BaseModel.
2. Provide build_runner_with_tta for using FSDP and TTA simultaneously. In this case, BaseTTAModel should be the wrapper of FSDPWrapper.
3. Add English and Chinese tutorial.

BC-breaking (Optional)

Does the modification introduce changes that break the backward-compatibility of the downstream repos?
If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR.

Use cases (Optional)

If this PR introduces a new feature, it is better to list some use cases here, and update the documentation.

Checklist

1. Pre-commit or other linting tools are used to fix the potential lint issues.
2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness.
3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMCls.
4. The documentation has been modified accordingly, like docstring or example tutorials.

[zhouzaida]

It would be better to update the pr message about which repositories have verified this PR and their PR links.