It is the author's conjecture that the local charge density on a Terrell rotated sphere will appear to change as more charged surface comes into view. This charge migration from the poles to the equator may also occur with relativistically rotating bodies thus increasing the apparent charge density orthogonal to the axis of rotation near the sphere's equator.
Since the SR raytracing algorithm is simulating the propagation of photons, insights into this phenomena should be forthcoming.
This may also be a result of relativistic residue. As a particle accelerates, its length in the direction of acceleration decreases due to Lorentz contraction, e.g. - a sphere becomes oblate. As the length decreases, the particle's charge, which is Lorentz invariant (the charge does not change with the relative speed of the particle) remains constant, hence the charge density increases (charge per unit volume increases). A charged test particle, lying in the plane perpendicular to the axis of acceleration of the accelerated particle, would be affected by this increase in charge density, as described by the Larmor formula (conjecture!). Hence the radiation seems to propagate in a two dimensional plane perpendicular to the direction of acceleration. Conjecture: this is the source of 'breaking' radiation.