I'm ultra close to the correct 2d order statistics. The tough part is that the normalization is not a constant, but changes for each term in the sum of P<x<y, P<x>y, P>x<y, and P>x>y. I figured that much out in the shower this morning.

For now I'll commit this. I want to make sure that those four P's are themselves
correct before I continue.

Changes to be committed:
modified: debinningCore/dataGen.py
modified: debinningMath/orderStatistics.py
modified: examples/nfrVrHDemo.py