[ALICE3] Add configuration indices to OnTheFly

[jesgum]

Change the configurable for the luts to be a vector where each element results in a new configuration that will be processed by the otf workflow. This means that each collision will have multiple instances of the same track (one for each config), and these can be separated by an index getter from a new table.

• All the luts provided will be loaded into memory
• If for some reason more luts are provided for one particle species than for others, when we "run out" of tables, we'll default to the first one provided.
• If only one configuration is used, everything should be remained unchanged and subsequent analysis tasks should run fine without requiring any change.

In general I'm not sure how good of an idea it is to have tracks from all different configurations to be associated with the same collision. @njacazio @ddobrigk please have a look and tell me what you think.

[github-actions]

O2 linter results: ❌ 80 errors, ⚠️ 90 warnings, 🔕 0 disabled

[jesgum]

I ran a local cross-check to see that if I only run one configuration, the current behaviour is well reproduced. I also checked
the performance with the current tracker run twice and the new one once with two configs.
See figures:

[jesgum]

@njacazio from my side I think that things should be ready to merge if you have no objections. In principle, more histograms from the tracker task should be moved to the dynamic histograms folders that are created but this can be changed in future pr's I believe. I made the change for the most important ones anyways

[njacazio]

Thanks @jesgum this is ok for the single configuration. The selection should be done everywhere.