Option to not sore per event scoring mesh histograms

[Stewart Boogert]

The root output files can be prohibitively large when storing a dense 3D scoring mesh. Some studies require only the complete run histograms and do not require the per event histograms, clearly this will limit the uncertainty calculation on the final run histograms if scoring quantities which are not Poisson (i.e counts of particles) e.g. energy deposit or dose.

The existing run histograms are created via the per event histograms and so the code needs to be untangled to allow and an option is required to disable per event histograms. Also need to confirm that the down-stream analysis codes are not impacted by the absence of the per event histogram.

‌