Some issues with `PauliTransferMatrix` and `ChiMatrix`

[akirakyle]

Currently the gates used in test_pauli,jl are not actually correct for their given names due to incorrect assumptions about "endiandness" (see qojulia/QuantumOptics.jl#426). For example

julia> q2 = SpinBasis(1//2)^2
[Spin(1/2) ⊗ Spin(1/2)]

julia> CNOT = DenseOperator(q2, q2, diagm(0 => [1,1,0,0], 1 => [0,0,1], -1 => [0,0,1]))
Operator(dim=4x4)
  basis: [Spin(1/2) ⊗ Spin(1/2)]
 1.0  0.0  0.0  0.0
 0.0  1.0  0.0  0.0
 0.0  0.0  0.0  1.0
 0.0  0.0  1.0  0.0

when CNOT should be

julia> q1 = SpinBasis(1//2)
Spin(1/2)

julia> CNOT = dm(spinup(q1))⊗identityoperator(q1) + dm(spindown(q1))⊗sigmax(q1)
Operator(dim=4x4)
  basis: [Spin(1/2) ⊗ Spin(1/2)]
 1.0+0.0im  0.0+0.0im  0.0+0.0im  0.0+0.0im
 0.0+0.0im  0.0+0.0im  0.0+0.0im  1.0+0.0im
 0.0+0.0im  0.0+0.0im  1.0+0.0im  0.0+0.0im
 0.0+0.0im  1.0+0.0im  0.0+0.0im  0.0+0.0im

Also this line in pauli.jl in the function that converts from ChiMatrix to PauliTransferMatrix hardcodes iteration over 2 qubit paulis so I think it will be incorrect for superoperators over other numbers of qubits (see here for the equation that this function implements). The test file only contains tests on 2 qubit superoperators so it wouldn't catch this.