Kirk, you have the essentials written down here very well.
The actual situation is as you describe, but swell starts from zero and could go up through as much as 0.2 mm not just 0.05 mm, and therefore you need to integrate dV/dT or volume with respect to time, and dC/dT or concentration wrt time. These sets of integrals describe difffusion in and out of the coating and therefore have both plus and minus values. There are several other terms that must be used in this type of evaluation, but it would be useless to continue with this here.
I was conservative in silver spread as well. It may be that a given film has twice the silver coated than I describe above.
In essence, in a dilute developer where the buffer capacity is low, or in a hydroxide developer where the buffer capacity is low, the pH drops and begins to contribute to edge effects in two ways. One is bromide ion effects and the other is hydrogen ion effects. It is difficult to distinguish between them.
The ascorbate, HQ, metol, whatever developing agents.. are present in small quantities mole wise to start with and can slow development by further dilution. Very often, they recycle by using the sulfite as an exchange medium, oxidizing sulfite to sulfate, or in the case of quinone, reacting directly to form the HQ monosulfonate. In any event, developer exhaustion can also contribute to edge effects. Interesting? I think so.
Therefore, all the obfuscation aside, your question is quite valid and with merit. The buffer capacity, not the pH may be quite significant in these very dilute developers, and is exaggerated by stand and semi-stand development. Developer concentration and ratios of the developing agents may be critical.
Since bromide drag is known to take place, we know diffusion is a significnat contributor to this situation. The difference is that the pH effect is more of a spread than a drag, ie, it spreads out more uniformly due to the diffusion effects of hydrogen ion being faster and more uniform. It is also more subtle. This also becomes a part of the integrals mentioned above.
Just as an aside, in color materials pH effects layerwise are an additional source of interimage effects due to alkali consumption of adjacent layers 'sucking' up or 'pushing out' byproducts such as hydroxide ion and hydrogen ion respectively.
This work has been published by researchers such as Kai Liang and Roland Willis in separate articles describing diffusion models and kinetics of development.
I apologize to all who don't like science, but sometimes it is good to get a healthy dose of reality to go with the 'black magic' that some people supply regarding photographic processes. It is a real science where math, chemistry, and physics meet head on.
The bottom line is that the dilute developers are on the verge of alkali starvation and developer starvation and this may introduce unpredictable effects akin to halide drag. This will vary from film to film due to halide content and silver content.
Careful examination of these types of developers and dilutions may reveal unique optimum formulas for given conditions of time, agitation or film type.
I hope this is a suitable explanation and help for you all.
PE