feat(docker_image): use cuda:12.3.0 base image

[xmfcx]

Description

Closes:

• Use a common base image for CUDA libraries #2446

Tests performed

Not applicable.

Effects on system behavior

Not applicable.

Pre-review checklist for the PR author

The PR author must check the checkboxes below when creating the PR.

• I've confirmed the contribution guidelines.
• The PR follows the pull request guidelines.

In-review checklist for the PR reviewers

The PR reviewers must check the checkboxes below before approval.

• The PR follows the pull request guidelines.

Post-review checklist for the PR author

The PR author must check the checkboxes below before merging.

• There are no open discussions or they are tracked via tickets.

After all checkboxes are checked, anyone who has write access can merge the PR.

[xmfcx]

If these jobs are successful, we can merge this.

• https://github.com/autowarefoundation/autoware/actions/runs/6879506877
• https://github.com/autowarefoundation/autoware/actions/runs/6879508781

[xmfcx]

@oguzkaganozt it seems both on arm64 and amd64 it built correctly. Could you review this PR?

Also they used the cuda base image as shown here: https://github.com/autowarefoundation/autoware/actions/runs/6879506877/job/18711567150#step:5:801

According to here:

autoware/.github/actions/docker-build-and-push/action.yaml

Line 90 in d3cd823

push: ${{ github.event_name == 'schedule' || github.ref_name == github.event.repository.default_branch || github.event_name == 'push'}}

It won't push it to the registry unless it was a scheduled job and was on the main branch.

But since it built things correctly, it should be alright.

[oguzkaganozt]

I think although it is fine to use the Cuda base image, maybe it is more convenient to use the Ros base image both for cuda and no-cuda images, Imho, this will simplify the base image logic. BTW, even if we use the cuda base image we would still need to download additional cudnn,cusparse, cublas etc. libraries. So using cuda-base does not eliminate the need to download cuda dependencies with ansible.

[xmfcx]

Related issue:

• Re-organize the Autoware Docker Containers #3999