[PWGUD] Fix global forward tracks selection and store tan lambda information

[diana0x0f]

The global forward tracks (MFT-MCH-MID type) were not selected properly for each BC, now this is done using the global BC info.
The tangent lambda information is stored in a new table to be able to perform the global muon refit as is done in DQ.

[alibuild]

Error while checking build/O2Physics/o2 for 1657ae4 at 2025-01-03 16:03:

## sw/BUILD/O2Physics-latest/log
/sw/SOURCES/O2Physics/9185-slc9_x86-64/0/PWGUD/TableProducer/UPCCandidateProducer.cxx:1620:15: error: unused variable 'nMIDs' [-Werror=unused-variable]
/sw/SOURCES/O2Physics/9185-slc9_x86-64/0/PWGUD/TableProducer/UPCCandidateProducer.cxx:1616:56: error: comparison of integer expressions of different signedness: 'const long unsigned int' and 'const long int' [-Werror=sign-compare]
ninja: build stopped: subcommand failed.

Full log here.

[rolavick]

Why do you introduce this new table? Apparently, there is already a table with this parameter here

O2Physics/PWGUD/DataModel/UDTables.h

Line 456 in a5e761a

fwdtrack::X, fwdtrack::Y, fwdtrack::Z, fwdtrack::Phi, fwdtrack::Tgl,

Another approach how to obtain track.tgl is to use the px, py, px information from

O2Physics/PWGUD/DataModel/UDTables.h

Line 396 in a5e761a

DECLARE_SOA_TABLE(UDFwdTracks, "AOD", "UDFWDTRACK",

which are actually calculated based on tgl see here https://github.com/AliceO2Group/AliceO2/blob/64dd90cfb914805f3334aa32022581e7a05971b7/Framework/Core/include/Framework/AnalysisDataModel.h#L785

[diana0x0f]

Hi @rolavick ,
we do not store the UDFwdTracksProp table, but sorry, I did not realise I could just use the px, py, pz information that is stored as a part of the UDFwdTracks table to calculate it. I will remove the new table.
Thanks for your help!