Add filter on the number of sim hits in the tagger and add detectors with just tracker & scintillator

[bloodyyugo]

We need to study the effect of off-energy beam electrons in our detector, namely how often they are identified as full-energy electrons. The probability for this is very small so we need to add a filter after simulation to reject events that clearly wouldn't be mistaken for full energy, namely they don't leave enough hits in the tracker and no hits in layers closest to target.

Omar had already done a study for 4gev electrons, see his presentation here, and has a filter but it was never committed. That's what I'm doing here.

In addition, we are adding 4 and 8 gev detectors with no ECal or HCal (just tracker and scintillator). Removing the calorimeters greatly speeds up simulation.