update base image to support cuda12.4 in dockerfile

[RunningLeon]

Motivation

update base image to cuda12.4

requested in #2164

Modification

Change dockerfile base image and support cu118 and cu124

build cu12

cd ~/lmdeploy
docker build . -f docker/Dockerfile -t lmdeploy:cu12 --build-arg CUDA_VERSION=cu12

build cu11

cd ~/lmdeploy
docker build . -f docker/Dockerfile -t lmdeploy:cu11 --build-arg CUDA_VERSION=cu11

docker build ci example

https://github.com/RunningLeon/lmdeploy/actions/runs/10211052906

BC-breaking (Optional)

Does the modification introduce changes that break the backward-compatibility of the downstream repositories?
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 tests to ensure the correctness.
3. If the modification has a dependency on downstream projects of a newer version, this PR should be tested with all supported versions of downstream projects.
4. The documentation has been modified accordingly, like docstring or example tutorials.

[lvhan028]

In the build-docker workflow, can we build openmmlab/lmdeploy:{version}-cu11 and openmmlab/lmdeploy:{version}-cu12

[zhulinJulia24]

by using image nvidia/cuda:12.4.1-devel-ubuntu22.04

leads to error like:
python3 -m pip cache dir
/usr/bin/python3: No module named pip

[RunningLeon]

by using image nvidia/cuda:12.4.1-devel-ubuntu22.04

leads to error like: python3 -m pip cache dir /usr/bin/python3: No module named pip

run it install python and create python venv

RUN rm /etc/apt/sources.list.d/cuda*.list && apt-get update -y && apt-get install -y software-properties-common wget vim &&\
    add-apt-repository ppa:deadsnakes/ppa -y && apt-get update -y && apt-get install -y --no-install-recommends \
    rapidjson-dev libgoogle-glog-dev gdb python${PYTHON_VERSION} python${PYTHON_VERSION}-dev python${PYTHON_VERSION}-venv \
    && apt-get clean -y && rm -rf /var/lib/apt/lists/* && cd /opt && python3 -m venv py3

ENV PATH=/opt/py3/bin:$PATH

[RunningLeon]

In the build-docker workflow, can we build openmmlab/lmdeploy:{version}-cu11 and openmmlab/lmdeploy:{version}-cu12

ok,

[zhulinJulia24]

by using image nvidia/cuda:12.4.1-devel-ubuntu22.04
leads to error like: python3 -m pip cache dir /usr/bin/python3: No module named pip

run it install python and create python venv

RUN rm /etc/apt/sources.list.d/cuda*.list && apt-get update -y && apt-get install -y software-properties-common wget vim &&\
    add-apt-repository ppa:deadsnakes/ppa -y && apt-get update -y && apt-get install -y --no-install-recommends \
    rapidjson-dev libgoogle-glog-dev gdb python${PYTHON_VERSION} python${PYTHON_VERSION}-dev python${PYTHON_VERSION}-venv \
    && apt-get clean -y && rm -rf /var/lib/apt/lists/* && cd /opt && python3 -m venv py3

ENV PATH=/opt/py3/bin:$PATH

env path is not set, fixed

[lvhan028]

LGTM

[zhulinJulia24]

lgtm