Draft: try including MC processing in generic RD production [to be finished/checked]

• Have a minimal example working at least for K JPsi/K* J/psi [K* J/Psi[mumu] MC ]
• the info.yaml has to do HLT1->HLT2-> Tupling [ currently SPRUCE is included, but not really needed]
• Interface has to have a list of 'known_event_types' for which a list of lines to run in hlt2.py step has to be known. We map HLT2-Line algorithms to HLT2-Line tupling, but we can also patch the HLT2 to load 'all lines to run in Moore' and run only the ones mapping to event type and supplied as extra argument in the AP.
• once HLT1->HLT2 runs good for an example, tuple it and check the MCTruth variables are created.
• MCDecayTree: see !1288 (diffs)
• Flavor Tagging following Clair's example: lhcb/DaVinci@18fc0781
• Backport K*gamma, Phigamma work for generic functors
• On X J/Psi channels text issues need to be fixed in the DTF maker
• Leon's comments: !1298 (comment 8022103)
• Add HLT2 Topo lines for Nicole's checks
• Add MC for B2HH, B2HH', B2MuMu, B24Mu
• No prio: Add isolation variables to B2LL and B2LLLL lines
• Add lines Hlt2RD_XibmToXimMuMu_DDD, Hlt2RD_XibmToXimMuMu_DDL, Hlt2RD_XibmToXimEE_DDD, Hlt2RD_XibmToXimEE_DDL, Hlt2RD_OmegabmToOmegamMuMu_DDD, Hlt2RD_OmegabmToOmegamMuMu_DDL