Switches to using before / after script for docker cleanup around each job
Work includes:
- Switching to the before + after scripts around each job to clean up the container images created.
Why do this?
Found that our runners have been running out of memory quickly. Looking on the runners, realize it is the ~40GB of container images left over. This should fix the xolotl-ci ones:
root@rse-devops-blue-gitlab-ci-runner-0:/var/lib/docker# docker image ls
REPOSITORY TAG IMAGE ID CREATED SIZE
code.ornl.gov:4567/code-fusion/xolotl-ci/develop-clang-asan-asan latest b1dd1a0e6bf9 10 hours ago 2.87GB
code.ornl.gov:4567/code-fusion/xolotl-ci/develop-source latest e768e62d810d 11 hours ago 2.87GB
code.ornl.gov:4567/sns-hfir-scse/sans/sans-backend/next latest d68e0a578a84 14 hours ago 5.35GB
code.ornl.gov:4567/code-fusion/xolotl-ci/stable-clang latest 8c8de247e87f 34 hours ago 3.68GB
code.ornl.gov:4567/code-fusion/xolotl-ci/stable-source latest 8252be380ee1 35 hours ago 2.86GB
code.ornl.gov:4567/sns-hfir-scse/sans/sans-backend/next <none> 14f8918d97b3 38 hours ago 5.35GB
code.ornl.gov:4567/sns-hfir-scse/sans/sans-backend/sans896 latest c1cc4bbbe2d5 2 weeks ago 5.31GB