Update docker docs for VL api

[vody-am]

Docs update in relation to issue #1514

Motivation

Updating documentation around Docker images for new users.
It should be made clear to users what is expected, and a way to resolve issues described.

Modification

Adding a warning about dependencies and an example of how to resolve by creating a new Docker image.

[lvhan028]

@vody-am If you are familiar with Chinese, could you please update the user guide in docs/zh_cn/serving/api_server_vl.md?

[vody-am]

@lvhan028 Google translate has been working well, so yes I will give it a shot 😂

[vody-am]

@lvhan028 while I am here -- is there anything special one has to set to use multiple GPUs? Thus far I have experimented with one container per GPU, but it seems like there should be a way to use multiple cards. Is it the --tp setting?

[lvhan028]

Yes, by setting tp parameter

[lvhan028]

Hi, @vody-am
there are linting errors, which are able to fixed by running the following commands:

pip install pre-commit
cd the/root/path/of/lmdeploy
pre-commit install .
pre-commit run --all-files

[zhyncs]

In a real production environment, the deployed model path is usually stored on something like Open Storage Service. Almost never download and load from HuggingFace every time you deploy.

[zhyncs]

Therefore, this is more of a simple reference and has little significance in actual production environments.

[vody-am]

Yes, that's a fair assessment.