[g13] fiducial cuts

[Paul Mattione]

I've changed the fiducial cuts for the g13a dataset: I've  
significantly expanded the back angle coverage by fitting the back  
angles with quadratics (instead of just a flat cutoff when the phi  
range decreases).  This should expand the angle range for all cross  
section calculations.  Also, these cuts are to first-order good for  
g13b as well, unless there was a different magnetic field strength...

Link (histograms, details on how to use the cuts):
http://clasweb.jlab.org/rungroups/g13/wiki_secure/index.php/FiducialCuts

  - Paul