AMH-115: Fix EXOT27 b-tagging in pflow and GhostTag VR track jets

The AntiKtVR30Rmax4Rmin02TrackGhostTagJets and AntiKt4EMPFlowJets were both pretty broken since they didn't include a b-tagging link and we weren't saving the corresponding btagging object. So we wanted to do a few things:

• Add b-tagging back
• Add a lot more variables that will be inputs to an Xbb tagger
• Keep the derivation the same size or smaller

So I've done a few things here:

• Add the b-tagging object to jet collections
• Schedule pflow b-tagging in this derivation (it wasn't being built before)
• Used a smart slimming list for GhostTag VR track jets that includes an additional 20 or so Xbb taging input variables.
• Used a smart slimming list for the GhostTag jets, previously we were adding them to AllVariables. This list is just the bare jet collection, I don't think we need anything else.
• Thin the GhostTag jets and associated b-tagging objects with @jburr's awesome link thinning tool.
  • I only save these jets if they are associated to a Large-R jet over 100 GeV
  • I'd consider going higher here but probably only if the we're thinning on are calibrated

The smart slimming list together with the added b-tagging variables means that we're actually saving slightly fewer variables per jet than we were before. In addition the link thinning reduces the number of jets we save to 1/3rd in the test I looked at (whatever the derivation framework uses for an MC file).

Also pinging @sargyrop, @eskorda. I think this is pretty much final but again we should run a few more tests before we merge / request a new cache.