Change units of asymmetries and differences to be in "convenient" units

[paulmking]

Describe the solution you'd like
Kent requests having all asymmetries in ppm and differences in microns. This could be done in JAPAN or in post-processing. Or just in the aggregator, or in the tree output routines.

Describe alternatives you've considered
Provide the "unit" in the root tree, so that if we plot "value/PPM" it is scaled to be in ppm and the plot has a indication of what unit is being plotted.

[wdconinc]

I think this was implemented in PR #122.