EPOS MB insufficient statistics for strange and light hadron analyses

An EPOS MB sample does not have enough statistics of reconstructed strange and light hadrons to perform an analysis.

This was observed in SMOG pAr 2024.W32.34 for the study of Lambda baryons (Sim10) from @mastefan : https://indico.cern.ch/event/1495957/contributions/6301529/attachments/3043616/5377289/simulationLambda_issue.pdf .

It has been previously observed for pPb EPOS simulations (Sim09). For example, in a MB sample with ~5M events, ~3M phi-1020 mesons are generated, of which ~600k are generated in acceptance and ~300k reconstructed.

LFN: /MC/2013/pPb-Beam4000GeV-1580GeV-2013-MagDown-Fix1-Epos/Sim09e/L0Trig0x1710/Trig0x406a1710/Reco14r1/Stripping20r3p1/30000000/ALLSTREAMS.DST

Suggested solution: Filtering at the generator level of EPOS MB events with strange and other light hadrons in LHCb acceptance.