Moller skimmer for 2019

[mgignac]

Implement Moller skimming, largely based on the V0 skimmer and adapted to Moller selection.

Local testing yields a filter efficiency of 2.4%:

Moller skim events processed = 500
Events passed            = 12
       Pass efficiency   = 0.024
Total number of Mollers processed = 40
Total number of Mollers passing skim = 14

[bloodyyugo]

Looks good to me.

[MYOMAO]

I tested locally at JLab ifarm with the steering file: PhysicsRun2019_pass1_recon_skimmed.lcsim on an evio. I got the data_events_mollerskim.slcio and data_events_v0skim.slcio in addition to the unskimmed data_events.slcio and the DQM root file: data_events_data_quality_plots.root. So I think it will also work for large scale swif2 job submission.