Fix {Pauli,SparsePauliOp}.apply_layout to raise an error with negative or duplicate indices

[t-imamichi]

Summary

{Pauli,SparsePauliOp}.apply_layout currently allow negative or duplicate indices.
The behavior is not always comprehensive.

1. If we give duplicate indices, some coefficients or phases may change.
2. If we give negative indices that are out of bounds, it raises IndexError while positive indices raises QiskitError.

This PR raises QiskitError with duplicate indices or negative indices (even within bounds for simplicity).
I think duplicate indices are not valid as a layout. But, if we need to support the pattern, we might need to update the code not to change the coefficients.

I noticed this situation when I review #12221 (comment)

from qiskit.quantum_info import Pauli, SparsePauliOp

op = Pauli("XY")
op2 = SparsePauliOp("XY")

print("duplicate indices")
try:
    print(op.apply_layout([0, 0]))
except Exception as ex:
    print(ex)
try:
    print(op2.apply_layout([0, 0]))
except Exception as ex:
    print(ex)

print("\nnegative indices not out of bounds")
try:
    print(op.apply_layout([-1, 0]))
except Exception as ex:
    print(ex)
try:
    print(op2.apply_layout([-1, 0]))
except Exception as ex:
    print(ex)

print("\nnegative indices out of bounds")
try:
    print(op.apply_layout([-3, 0]))
except Exception as ex:
    print(ex)
try:
    print(op2.apply_layout([-3, 0]))
except Exception as ex:
    print(ex)

print("\npositive indices out of bounds")
try:
    print(op.apply_layout([3, 0]))
except Exception as ex:
    print(ex)
try:
    print(op2.apply_layout([3, 0]))
except Exception as ex:
    print(ex)

main

duplicate indices
-iIX
SparsePauliOp(['IX'],
              coeffs=[0.-1.j])

negative indices not out of bounds
YX
SparsePauliOp(['YX'],
              coeffs=[1.+0.j])

negative indices out of bounds
index -3 is out of bounds for axis 1 with size 2
index -3 is out of bounds for axis 1 with size 2

positive indices out of bounds
'Provided layout contains indices outside the number of qubits.'
'Provided layout contains indices outside the number of qubits.'

this PR

duplicate indices
'Provided layout contains duplicate indices.'
'Provided layout contains duplicate indices.'

negative indices not out of bounds
'Provided layout contains indices outside the number of qubits.'
'Provided layout contains indices outside the number of qubits.'

negative indices out of bounds
'Provided layout contains indices outside the number of qubits.'
'Provided layout contains indices outside the number of qubits.'

positive indices out of bounds
'Provided layout contains indices outside the number of qubits.'
'Provided layout contains indices outside the number of qubits.'

Details and comments

[qiskit-bot]

One or more of the the following people are requested to review this:

• @Qiskit/terra-core

[jakelishman]

Thanks, this seems like a good idea to me. Could you add a brief bug-fix release note mentioning the change?

[coveralls]

Pull Request Test Coverage Report for Build 9030840979

Details

• 8 of 8 (100.0%) changed or added relevant lines in 2 files are covered.
• 23 unchanged lines in 2 files lost coverage.
• Overall coverage decreased (-0.03%) to 89.626%

---

Files with Coverage Reduction	New Missed Lines	%
crates/qasm2/src/lex.rs	5	92.88%
crates/qasm2/src/parse.rs	18	96.69%

Totals
Change from base Build 9023909750:	-0.03%
Covered Lines:	62201
Relevant Lines:	69401

---

💛 - Coveralls

[jakelishman]

Looks good, thanks for the quick change and the fix!