I've been poking this for hours, and I have no idea why the current implementation of QStabilizerHybrid::Clone() doesn't seem to work when copying a QStabilizer pointer. However, I will open an issue to track this, and it shouldn't make any significant difference to performance, since this case is invoked in Qiskit only immediately before a non-unitary gate, as in the unit test.
I've been poking this for hours, and I have no idea why the current implementation of
QStabilizerHybrid::Clone()
doesn't seem to work when copying aQStabilizer
pointer. However, I will open an issue to track this, and it shouldn't make any significant difference to performance, since this case is invoked in Qiskit only immediately before a non-unitary gate, as in the unit test.