As others commented here, a 300-600, be it f/4.0-5.6 or constant f/5.6, would need a big feature that really sets it apart from the 100-300 f/2.8 + 2.0x TC. For real life photography, that's weight. Let's speculate that Canon may introduce diffraction optics (DO) in the first supertele zoom on the market to make it extremely light. To my knowledge, all tele lenses with DO/PF features so far made are primes: Canon's pioneering 400 f/4.0 I & II, Nikon's 300/4 that came later for their F mount (and suffered initially from decentered lenses, Nikon's production was still in a learning curve), and now Nikon's bold and great move with the 600 and 800 f/6.3 "PF" lenses. Let's look at the physics: diffraction optics means to use Fresnel lenses, flat lenses designed with tiny concentric rings that diffract light basically like a much more massive conventional glass lens (in fact a set of conventional glass lenses). This allows for a very light, compact lens design, but the price is that a sort of micro pattern comes into play. I guess that's the reason why we do not yet see any zoom lens based on DO (please correct me if I missed any lens on the market, Nikon's 180-600 is "non-PF", I checked that). So I guess it is a real challenge to control any sort of optical effects created by such micro pattern well enough for a high-end tele zoom.
That DO may create problems came into my wife's (she is a physicist) and my mind during a birding trip from which we returned yesterday. We shot birds side by side, and if there was enough light with the same settings. My wife used her new Z8 (upgraded from a Z5 II), her 600/6.3 PF lens with and w/o Nikon's 1.4x TC for Z mount. I used my EF 600mm f/4.0 III, w & w/o 1.4 TC III, and mostly my R5 II (partly my R7 when there was enough light). Without TC's my wife's combo performed AF wise like my combo (aperture set to f/6.3-8.0 to gain more depth of field): fast & precise. But with the 1.4x TC on, my wife's Z8 started to struggle: it's AF found the bird quickly but then started to micro-pump around the precise in-focus distance. Only if the bird was sitting still for a quite long time, the camera finally found the correct in-focus distance by try & error, just like the old purely contrast-based AF systems. By contrast, my Canon combo worked fast & precisely. But what is the problem in the Nikon system? One source could be Nikon's TC, of course we could have purchased a bad copy. But with the Z5 II my wife had not such bold problems with this TC, so we came to the conclusion the TC isn't the bottleneck. Now, the Z5 II has only 24 MP, so it has much less resolution and a much "grainier" pixel pattern on its sensor. The Z8 is known for being prone to Moiré, Nikon followed their philosophy for sacrifying a stronger optical low-pass filter in front of the sensor for gaining more micro-contrast. So, currently we suspect that the pattern of the diffraction optics in the 600/6.3 PF may interfere in some way with the Z8's sensor when magnified by the TC. But that's only our current guessing. One way to test it would be to use a comparable lens with conventional optics, what means that we would have to rent a Z 180-600 zoom - which is known to be much lesser sharp with a TC.
Well, that has to wait until we have time again to deal with that problem, since now we are back to work again. I also will try to figure out the "pitches" of such DO lens rings compared with the pixel pitch of such a 45 MP FF sensor: are these two distances even in the same order of magnitude, and is there any pattern to be expected on the optical image plane on the sensor? In fact, it needs to interfere with the pixels used by the AF system on Nikon's sensor, so we would need this information...
Sorry for this lengthy posting, but maybe someone has a good idea here regarding our "DO problem".