Discard afterburner tracks

[shahor02]

https://its.cern.ch/jira/browse/QC-1114

[github-actions]

REQUEST FOR PRODUCTION RELEASES:
To request your PR to be included in production software, please add the corresponding labels called "async-" to your PR. Add the labels directly (if you have the permissions) or add a comment of the form (note that labels are separated by a ",")

+async-label <label1>, <label2>, !<label3> ...

This will add <label1> and <label2> and removes <label3>.

The following labels are available
async-2023-pbpb-apass
async-2023-pp-apass1
async-2022-pp-apass6
async-2022-pp-apass4
async-mc
async-data

[f3sch]

Hi @shahor02, intuitively I would have thought this should be added here

 for (int itrk = 0; itrk < static_cast<int>(mITSTPCTracks.size()); ++itrk) {       // line 461
      auto const& trk = mITSTPCTracks[itrk];
      auto idxTrkTpc = trk.getRefTPC().getIndex();
      if (trk.getRefITS().getSource() != GID::ITS) {
             continue;
        }
      if (isTPCTrackSelectedEntry[idxTrkTpc] == true) {
        auto lbl = mRecoCont.getTrackMCLabel({(unsigned int)(itrk), GID::Source::ITSTPC});
        if (!lbl.isValid()) {
          continue;
        }

so before we build this mMapLabels, because otherwise I could imagine that this would somewhat mess with the normalisation. In this case we would exclude AB tracks entirely from the efficiency (might also not be ideal). I tested that they stop the crash.

[shahor02]

@f3sch Makes sense, thanks!

[chiarazampolli]

I added the labels for pp 2023 apass3/PbPb apass3 only.

[chiarazampolli]

Tested by @f3sch .