Fix for Pythia8 UserHook to enable decay to SUEP
This MR addresses an issue that was found only with large-stat validation of the code introduced in !34799 (merged).
In rare circumstances, when too few particles are produced in the shower, the energy conservation fails and the code crashes badly. We protect against this rare occurrence in this update.
Once this update is accepted, we'd need a new release including it (apologies for the extra work). It would be important that the Pythia version is the same as for 21.6.37 (or please let me know so I can perform adequate tests).
A MC request using this code was approved by group convener and is currently on hold waiting for this fix (https://its.cern.ch/jira/browse/ATLMCPROD-8730).
Tagging @mehlhase as well.
Thanks and apologies for this extra iteration!