FifoPlayer: Wait after clearing the screen

[Pokechu22]

Continues #10258. If we don't wait, then the copy will be performed at a later time, which may overwrite memory updates. This was causing problems for me when messing with memory updates, since the EFB copy would end up overwriting memory that had been written by the memory update. I think this may have been the cause of some of the differences in #10264, and will submit a second PR to check.

[Pokechu22]

#10274 shows that this improves early memory updates (though early memory updates still isn't a good solution). So although this produces no visible fifoci differences, it is an improvement.

[iwubcode]

I ran some fifo logs and all of them ran as a black screen. Master shows them correctly.

[Pokechu22]

It seems like that happens with dual core (but not single core). However. fifoci uses dual core if I recall correctly, and it ran without issues so I'm not sure quite what's going on.

[Pokechu22]

Fixed. I needed to add this to the fifoplayer's CPUCore::Init:

dolphin/Source/Core/Core/FifoPlayer/FifoPlayer.cpp

Lines 217 to 219 in 86f2b39

	// Without this call, we deadlock in initialization in dual core, as the FIFO is disabled and
	// thus ClearEfb()'s call to WaitForGPUInactive() never returns
	CPU::EnableStepping(false);

(Also, I was incorrect in my thought that FifoCI uses dual core; I misinterpreted what CPUThread = False meant in configs. CPUThread = False actually means "don't use a separate CPU thread from the GPU thread", i.e. "don't use dual core".)

[iwubcode]

Dual core and single core now work. Code changes make sense