Add a check if stress initialisation is actually necessary to save memory
We had a couple of HBM simulation runs which produced binouts with a combined size of 26 - 36 GB. We were only able to analyse the results with Dynasaur after we commented out the line "self._assign_stresses(elout_elem)" in the file "ls_dyna.py". This is not needed if no stresses are evaluated. Before this fix, we always ran out of memory. Ideally, Dynasaur should check if stresses are acutally relevant and only use memory to read them in if that is the case.
Regards, Felix