Two reasons are given to explain the black drop effect here, but I think I came up with a third.
Consider a two-dimensional cross section of the situation just after second contact and just before third contact in a stellar transit.
It has many similarities to the classical single-slit experiment:
a b
============================|---|====================|-----------------------
Empty Space ↑ Disk of Planet Disk of Star
|
Edge of Disk of Star
Our slit begins at point a and ends at b.
Consider that the diffraction pattern of the single-slit experiment spreads out the total energy of the light which is incident on the range of the slit over some angle. This indicates that the incident intensity of the light behind a slit is less than the intensity would be if no slit was present (and no diffraction occurs).
Could this possibly help account for the Black Drop phenomenon? Or must the "slit" be a lot smaller, such that this could not possibly contribute in any significant way? After all, the aperture in this case would be on the order of hundreds of kilometers.
Another thing that just occurred to me is that the left-edge of the star is not being obscured by anything, only the right side is. There would only be one-sided diffraction affecting our incident light intensity.