I pre-wash to preserve my Replenished Xtol developer from accumulating antihalation dye, spectral sensitising dye. acutance enhancing, dye, and possibly speed trimming dye. A replenished developer loaded with dye becomes a murky mess of unknown (unknowable?) composition that prompts discard. By keeping dyes out of my Xtol it stays clear-ish and can do many many films without an alarming change of appearance.
Hi, fwiw I spent a lot of years as a large-lab QC guy, my department overseeing the "process control," results of chem mix, and loads of troubleshooting. When I started this as a near wet-behind-the-ears kid it was with the predecessors (C-22 and Ekt 2/3) of today's C-41 and RA-4. Also a handful of other processes, b&w, all replenished. None of these ever used a pre-wash, and never had any obvious problems related to lack of same. But... I have never dealt with Xtol other than my personal use.
There IS one significant difference in that every one of these machines used filtered recirculation systems (where said filters collected considerable "sludge"). Which I sorta presume the small-scale home users are most likely not using.
FWIW we pulled chemical samples, about 40 to 50 per day, from (all) of our processing machines for basic screening (pH and specific gravity). All of the color developers took on a color unique to the specific materials being processed (we kept them segregated on different processors). But all developers were "clear;" no obvious cloudiness. And, as I said, never any obvious problem related to build-up of whatever byproducts.
With respect to (non-development) byproducts building up in a developer, the concentration is ultimately limited by dilution due to the replenisher. Occasionally people on this forum will say that, "oh, if you don't periodically discard some volume of developer then the concentration of byproducts will just keep increasing." But this is not true. These concentrations are controlled by the amount_released per the volume of replenisher added. (Ignoring evaporation.) Whether developer is discarded or not is essentially irrelevant to this although, for practical purposes, some has to be discarded sooner or later.
FWIW the lowest replenishment rate system for a C-41 developer is Kodak LORR replenisher, which runs at a rate of approximately 25 to 30 ml/roll (135-36) as I recall. In other words all byproducts released (into developer) from a single roll of film will, loosely speaking, be contained in that 25-30 ml of added replenisher volume. (The equilibrium concentration will be: total byproduct divided by volume added.) Now, my understanding is that replenished Xtol gets replenished at something like 2 or 3 times that of the LORR. Which means that, all else being equal, replenished Xtol should only reach something on the order of 1/2 to 1/3 of the (non-development) released-component concentration that a C-41 developer would have. Which in my C-41 experience is never a problem. So I would tend to suspect Xtol would likely not have a problem either. But I don't have any first-hand knowledge of this. And I am glossing over some of the details. So this is just some food for thought.