Multi-Control Pauli Decomposition is probably incorrect

[tanujkhattar]

Right now, we don't have a test for correctness and as part of writing a gate locally, I discovered that the gate is probably not doing what it's promised to do (I tested for cirq.Z(q).controlled_by(*control_qubits, control_vals = [0] * len(control_qubits)))

This issue is to verify the correctness and add a validation test:

https://github.com/quantumlib/cirq-qubitization/blob/a2f283e0269d92d89aba2ce19e5b973019db5aca/cirq_qubitization/cirq_algos/multi_control_multi_target_pauli.py#L44

[fdmalone]

@tanujkhattar I hit this recently and remembered I was supposed to look into it. Any idea what's going on before I dig in?

[mpharrigan]

@tanujkhattar did you fix this?

[anurudhp]

#790 added tensor tests to verify the action, closing.