Docker labels derived from env variables set once per dependency

[bryngemark]

I am changing stuff in the Dockerfile, here are the details.

What new packages does this PR add to the development container?

None. Instead it makes sure we define labels based on the same version number variable as we use for building the container. This closes #77 .

Check List

• [x] I successfully built the container using docker

  # outline of container build instructions
  cd docker
  git checkout origin iss77
  docker build . -t ldmx/local:dockerIss77

• [x] I was able to build ldmx-sw using this new container build

  # outline of build instructions
  ldmx use local dockerIss77
  cd ldmx-sw
  mkdir build
  cd build
  ldmx cmake ..
  ldmx make install

• [x] I was able to test run a small simulation and reconstruction inside this container

  # outline of test instructions
  cd $LDMX_BASE/ldmx-sw/build
  ldmx ctest
  cd ..
  for c in `ls */test/*.py`; do echo -e "\n --------------- \nTesting with $c " ; echo ; ldmx fire $c ; done

• [x] I was able to successfully retrieve the version labels

  docker inspect ldmx/local:dockerIss77
  [...]
          "Labels": {
              "catch2.version": "3.3.1",
              "eigen.version": "3.4.0",
              "geant4.version": "LDMX.10.2.3_v0.5",
              "genie.version": "3.02.00",
              "lhapdf.version": "6.5.3",
              "onnx.version": "1.15.0",
              "org.opencontainers.image.ref.name": "ubuntu",
              "org.opencontainers.image.version": "22.04",
              "pythia.version": "6.428",
              "root.version": "6.22.08",
              "ubuntu.version": "22.04",
              "xercesc.version": "3.2.4"
          },
  [...]

[bryngemark]

I can get to the ldmx-sw building tests next week.

[bryngemark]

@tomeichlersmith there is a disk space issue preventing the CI build. Do we have a cleanup scheme for old containers, or are they always wiped between attempts?

[tomeichlersmith]

The old containers should be wiped but I manually removed the cache now. I am uncertain why there was still caches left, maybe due to the failed builds? I don't know.

[bryngemark]

Ok. Would it make sense to add whatever you did to clear the cache to the end of the CI workflow? It's awesome to profit from the cache to save on build time but if all we get is space problems, then it's not working for us.

[bryngemark]

Running the tests set up in the different modules uncovered a whole lot of issues -- but they look like the tests haven't been updated in a while. I'll make an issue. I did manage to successfully run e.g. the ecal PN test.

[tomeichlersmith]

The cache-clearing is already at the end of the CI workflow, I don't think it is called if the build fails and ends the workflow early, I need to look into if we can always clear the cache in some way.

https://github.com/LDMX-Software/docker/blob/93cad7ece7c84fc99f4f9a9bafb2192858ef90e6/.github/workflows/ci.yml#L65-L72