Open Dataset Cleanup

We have to get moving on #142.

There's a bit of work needed to clean up the dataset. I covered a few things here:

• I'm moving most of the "fragments" into the main configuration file so that it's easy to get a list of what we're doing / to increase coherence.
  • In the truth particle collection I noticed that we use displacement<X> and decayVertex<X> to mean the same thing. Went with decayVertex for consistency.
• We required a truth primary vertex match for some reason, I assume because enabling the matching also happened to enable the decorations giving the displacement between the two. I made the decorations and the selection independent.
• I tried to clean up the logic in processEvent a bit: don't read the same container multiple times, make the function names a bit more clear and specific.

I'm also removing some residual AnalysisBase stuff.