Add resource-group semaphore to force pipelines to run one after the other to avoid race conditions on the NFS mount

Transfer to the OpenShift pod that will perform the build, the script needed to create the configuration for the respective spackified JupyterLab kernel

Defined some global variables into CI/CD Variables so that a change in them does not result in new commits

Code cleanup and final touches for sending spack yaml files and packages to the Openshift build job through the Gitlab runner instead of using SWIFT as the intermediate

Propagate the logs of the OpenShift build job to the Gitlab runner so that pipeline watchers can see the build job logs

Modify gitlab-ci.yml to have a different simplejob per build in OpenShift so that we can keep the logs for each build process and do not delete previous ones