Simon Schuett authored 3 weeks ago and Volker Friese committed 2 weeks ago

Fix cluster and hit finding in the STS detector when using 32-bit floating-point
precision. The issue occurs when calculating time differences between digis or
clusters with large timestamps - 32-bit floats lack sufficient precision for
representing nanosecond timestamps in the mCBM setup.

Add helper methods GetTimeDiff() that first convert timestamps to int before
calculating the difference, preserving precision and sign. Replace direct
floating-point timestamp subtractions throughout the code with calls to these
helper methods.

This resolves hit count discrepancies between 32-bit and 64-bit float versions,
reducing the difference from ~2.6% to ~0.04%.

 - QA for V0-trigger
 - Masking channels (BMON as example) in digi event-selector

With newer version of macosx and/or XCode the Geant3 code compiles without
problems but when it comes to memory allocation during runtime the execution
crashes. Up to now the underlying problem couldn't be identified.
Change CI tests such that on macosx only Geant4 is used. On Linux still Geant3
is used as simulation engine.

Bugfix for 2-station CFV MVD which corrects number of sensors in 2nd station.

Hits across sensors for 2 station MVD.

Rational:
- was introduced because bare-metal OS on virgo and virgo2 is centOS + debian containers were not immediately available stable
- now both vae23 (debian10) and vae24 (debian11) are available and well established/used by CBM virgo users
- these two jobs were responsible for ~50% of all pipeline failures due to something hanging up in the load_libraries test at a cost of 2 x 900 s

kfp-online: linking ROOT::Hist and ROOT::MathCore, changed the dependency on CbmKFParticle<Online/Offline>Interface to PUBLIC

- using a KFParticle version with no ROOT dependencies