bin edges for growth rate integration are poorly defined

[JonathonMSmith]

Currently, there is not a rigorous approximation of the physical bin sizes for the flux tube cells, and the dimensions are taken as half the distance between the values of neighboring bins. This echoes the approach in matplotlib's pcolormesh but should not be the only option here.