Investigate options to improve Dockerfile

[elevran]

For example, optimize order of commands for layer caching, etc.

[elevran]

this task requires good understanding of docker and CICD, not the llm-d code base, hence marking as good-first-issue

[github-actions]

This issue is marked as stale after 90d of inactivity. After an additional 30d of inactivity, it will be closed. To prevent this issue from being closed, add a comment or remove the lifecycle/stale label.