Thanks yes, i think we would all tend to agree with that....except of course at light load where the CCM PFC would be in DCM anyway.....and the BCM would be switching at a few hundred kHz, and probably not getting nice "valley" switching like that.
...Though actually, the PFC is a non isolated stage, so there is less scope for Common mode emissions to wreak havoc with Boost PFC. (Compared to an isolated transformer based stage).
In my view, the best thing that can be done with a BCM PFC controller is convert it to a constant off time (C.O.T) controller, and just arrange the boost inductor etc so that you are just inside DCM at max load, minimum vin......then you will be in DCM during the RFI test....and since you have chosen C.O.T, you will not get such ridiculously high fsw at light load.
The L6564 BCM PFC controller's literature actually gives instructions on how to convert it to a C.O.T. controller (!).....they can see the bubble is bursting i think.
Of course, if you're on Maximum load all the time...then i would admit BCM PFC is best.