Reset statement wrapper

[johnzl-777]

#510 Brings back the Reset statement (as well as dropping ResetToOne) but this statement does not have a wrapper that conforms to the latest operator semantics.

I had tried in that PR to just refactor qubit to follow the format of the gate dialect but that broke enough old tests I felt it warranted a separate PR instead of being shoehorned into #510.

Note: there is a test in #547 that has a local wrapper for reset that should be removed once this is properly implemented (:

[johnzl-777]

I think if @david-pl knocks out #549 this will naturally be closed by it