[ALICE3] Update V0 reconstruction in onTheFlyTracker

[lhusova]

fix the code for V0s

[github-actions]

O2 linter results: ❌ 27 errors, ⚠️ 21 warnings, 🔕 0 disabled

[alibuild]

Error while checking build/O2Physics/o2 for e8c8e31 at 2025-12-01 07:06:

## sw/BUILD/O2Physics-latest/log
/sw/SOURCES/O2Physics/14015-slc9_x86-64/0/ALICE3/TableProducer/OTF/onTheFlyTracker.cxx:809:33: internal compiler error: in tsubst_expr, at cp/pt.cc:21462
ninja: build stopped: subcommand failed.

Full log here.

[jesgum]

Hi @lhusova! I think that this could be problematic for analyses that does not use the fastTracker such as the D0 and Lc. Maybe we could instead use the size of the lut vectors for the primary particles?

[lhusova]

Hi @jesgum, probably it would be the best to do two separate loops? for Secondaries I don't need LUTs and it make sense to me that the loop goes over the different geometries. If you think that for the primaries LUTs loop would be better, then this would be an option. What do you think?

[jesgum]

Hi @lhusova, I think that doing it as two loops would be OK. One just needs to be careful such that the index matches between the loop for the primaries and the secondaries.
Because in my analysis for example, I use both, so it's important for me that one index points to the same geometry in both cases so I don't use one geometry for primaries and a different one for secondaries

The process function would also have to be adjusted similarly then I think onTheFlyTracker.cxx#L1544

[lhusova]

Hi @jesgum, please have a look how I solved it. I don't think that the line 1544 needs a change as the smearer is created as many times as this loop.

[jesgum]

Many thanks @lhusova!

[vkucera]

@lhusova The PR title doesn't say anything about what was done.

[njacazio]

Thanks @lhusova