As discussed in the Feb. 23rd tracking CP plenary, let's move to a setup where the dcube xml definitions for the InDetART tests are tracked in git. It should be then easier to iterate on a set of histograms that minimizes false positives. In this MR I only set this up for a single test (test_run4_el10) to avoid breaking all the tests in the event that the method I chose does not work in the ART environment. If it works, we can then move the rest of the tests to such a setup.
In the same go, I commented out the hit-level plots since the required collections are not filled by the jobs at the moment. We are discussing whether or not we want to have them in the future, hence they were not deleted yet.