applyMemoryError() needs to be redesigned

[rdviii]

Right now, whenever you create a new function like Xpurify() that operates on qubits, you have to explicitly call ApplyMemoryError(). As of this writing, it isn't incorporated into entanglement swapping properly yet. This is rather ad hoc, and we want people to be able to add new circuits without having to do a lot

Instead, this really ought to be incorporated into the gate class; any time a gate gets called on a qubit.

The catch in entanglement swapping is error propagation.