Add ACTS-based full reconstruction flow (no filter)

[kkrizka]

Adds an example reconstruction flow that replaces the tracking with the ACTS processor. It targets events without BIB as there is no track filter to remove fake tracks. The filter is available in the unreleased TrackPerf package. It is added as 20-acts-nobib, with the idea being that 2#-* will correspond to ACTS configurations.

The ACTS tracking is added as an extra option to TrackingChoices. Selecting ACTS enabled the following algorithms:

• ACTSSeededCKFTrackingProc
• ACTSDuplicateRemoval

The output of ACTSDuplicateRemoval is the SiTracks collection that is then refit by RefitFinal. The RefitFinal is already part of the usual workflow for the other tracking configurations. Its purpose for ACTS is to create the track state at the calorimeter (needed for pandora). The ACTS processor currently does not do this.

Implemented by @Benzillaist presentation. Electron gen/sim taken from #7 .

[gianelle]

• Why this pull request is "merged" with the other pull request: MC electron gun: added the Pandora settings #7 ? Can we manage them separately?

• What are the differences between this configuration file and the ones commited here: https://github.com/MuonColliderSoft/ProductionConfig/tree/main/reconstruction/005-Acts (at least for what concern ACTS side)?

[kkrizka]

1. This PR is based on the steering file introduced in MC electron gun: added the Pandora settings #7 and reuses the gen/sim steps in the CI. I am happy to rebase and merge this after MC electron gun: added the Pandora settings #7 is done. Is there something preventing it from being merged?
2. The 005 is an example of how to use ACTS for tracking. This PR puts the ACTS tracking in the full reconstruction chain. Do you think this should replace 005 to reduce any confusion?

[gianelle]

1. No, I don't see any problem in merging it
2. I prefer to keep the 005 configuration because it has been used in many official reconstructions