With regards to the theory that the phosphors used in the LED backlight display of the rMBP panels is what is causing the image retention, I think that part of the confusion is that some posters are referring to the LCD as just the "liquid crystal" part, i.e. the layer that has the crystals in it and the associated electronics, as separate from the backlight part, whereas you (High-Death) are referring to the entire assembly (including backlight) as the LCD. I don't know if either way is more correct than the other, but I would like to point out that this is probably some of the confusion.
The reason that this distinction is important is that most people believe that the image retention is being caused by some persistent electric field that is built up over time by bright images (i.e. parts of the LCD where the crystals are being held in orientation that lets the most backlight through) and that then prevents the crystals from properly conforming to the correct orientation until the electric field dissipates. It seems like what is happening is that the built-up electric field doesn't prevent the crystals from reorienting, but it does ever so slightly attenuate their re-orientation, which results in them shifting less than (or more than, not sure what direction it is) they are supposed to, which results in slightly different brightness from those pixels than from the surrounding pixels that weren't subject to the built up electric field. Over time the electric field dissipates and the crystals end up responding to voltages identically to other pixels which then smooths out and eliminates the image retention.
I may have gotten some of the terminology wrong or some of the concepts backwards, but I think that's a reasonable summary of the theory that I've heard most often proposed as the possible cause of this.
This theory is entirely dependent on the liquid crystals themselves; as far as I know, it doesn't care about the backlight at all (unless the material that the backlight is made of is somehow part of what is causing the electric field ot build up? But if so, I don't think I've heard that proposed yet), so whether or not phosphors are used to generate the backlight or not is not relevant to the theory.
This is in contrast to your theory, which I don't fully understand but which I think somehow depends upon the phosphors in the backlight themselves. I don't actually understand how they could possibly matter though as the backlight is not enabled or disabled, or brightened or darkened, per-pixel. So if the phosphors are always equally excited while the backight is on, what would cause them to somehow later on operate differently for the portions of the screen with image retention?
The fact that nobody else has yet proposed that the backlight has anything to do with the image retention (aside from you, High-Death, if I read what you wrote correctly), is likely the reason that others have said that the phosphors have nothing to do with the problem - because in the prevalent theory of what could cause image retention, the backlight in fact has nothing to do with it. But this does miss the point of your theory, which is in fact that somehow the phosphors do have something to do with it, although like I said earlier, I don't really understand what you are proposing well enough to understand exactly how the backlight phosphors are involved.