[PWGCF,PWGEM,PWGLF] Replace expression columns with dynamic columns in strangeness data model

[romainschotter]

Replace expression columns to dynamic columns in the strangeness data model.
Within the strangeness framework, expression columns appear only in extended tables and, more particularly, in the tables V0Cores, V0fCCores, CascCores , KFCascCores and TraCascCores. In order to ensure backwards compatibility, the extended tables (V0Cores, V0fCCores, CascCores , KFCascCores and TraCascCores) have been replaced to aliases to standard tables (V0CoresBase, StoredV0fCCores, StoredCascCores, StoredKFCascCores and StoredTraCascCores respectively) and the previously existing expression columns have been added as dynamic columns to the aforementioned standard tables.
This has several consequences:

• as dynamic columns cannot be used for filtering, all Filter using one of the above tables have to replaced by imperative-if conditions,
• as there are no more expression columns, there is no need to spawn the extended tables and for spawner tasks anymore, thus allowing to reduce the memory consumption by 50MB.

In order to make sure that analysis results remain unchanged after those changes, here is a comparison of the invariant mass distribution of K0s in Pb-Pb at 5.36 TeV (with 1 file of run 544116) obtained using a O2Physics installation with the expression columns in the strangeness data model in full black circles and without them in open red circles.

As you can see, the results are exactly the same before and after the changes.

@ddobrigk, for your information

[github-actions]

O2 linter results: ❌ 839 errors, ⚠️ 0 warnings, 🔕 0 disabled

[ddobrigk]

Hi @romainschotter, thanks a lot for taking care of this! I hope you don't mind - I took the liberty of commenting inline (just for convenience).

For reference: the removal of the expression column spawning will reduce the memory footprint further by approximately 60MB, which, at a time in which memory in core services is a commodity, is a very good thing :-)

[romainschotter]

Regarding the changes in CF code, @JStaa (the author of the threeParticleCorrelations.cxx task) has reviewed and agreed with the modifications introduced in this PR.

[alibuild]

Error while checking build/O2Physics/o2 for fcc22e3 at 2025-06-02 17:50:

## sw/BUILD/O2Physics-latest/log
/sw/SOURCES/O2Physics/11420-slc9_x86-64/0/PWGCF/MultiparticleCorrelations/Tasks/threeParticleCorrelations.cxx:884:9: error: 'v0' was not declared in this scope; did you mean 'o2::aod::v0'?
/sw/SOURCES/O2Physics/11420-slc9_x86-64/0/PWGCF/MultiparticleCorrelations/Tasks/threeParticleCorrelations.cxx:886:18: error: 'v0' was not declared in this scope; did you mean 'o2::aod::v0'?
ninja: build stopped: subcommand failed.

Full log here.

[romainschotter]

@saganatt @victor-gonzalez @zchochul @lgraczykCern @prchakra @lauraser @ariedel-cern @EmilGorm @otonvd @shouqiye @glromane can you please have a look at the changes in the PWGCF directory and approve them if they look okay to you? @JStaa (the author of the threeParticleCorrelations.cxx task) has reviewed and agreed with the modifications introduced in this PR.
@dsekihat can you please re-approve the changes? Somehow, Github introduced an error when merging my fork with the master branch.

[victor-gonzalez]

Thanks @romainschotter!!!