21.3 ffabbri fix in det phys val monitoring tool

This merge request solves two issues that don't allow to run the fast digitization validation ATLPHYSVAL-634. The issues appears when storing both the pileup and the hard scatter truth but I think is a fundamental weakness in the code.

(*(matchedEProbs[deg_count]))++;

matchedEProbs has a fixed length of 5 but deg_count (number of tracks matched to the same truth) has no upper bound, so this was trying to access a not existing piece of memory. I also added a check on the link to truth particles that need to be not nullptr to be used.