Add display of internal circuits for IfElseOps and WhileLoopOps to mpl circuit drawer

[enavarro51]

Summary

This PR adds display of the circuits inside IfElseOps to the mpl circuit drawer

Details and comments

UPDATED: Now boxes properly fold onto the next line and no longer a WIP.

UPDATE 2: Added WhileLoopOps

This PR is intended to be merged after #10096. It should accomplish the following

1. First, do no harm. All circuits without IfElseOps should display EXACTLY as they do in main.
2. Whether created directly as IfElseOp or through qc.if_test all if-else should display with a box with the if and else circuits inside.

Testers Wanted!

Since the circuits inside the if-else can be very large and complex, it's difficult to test all the scenarios. If you have a qiskit-terra development environment on your local system, just create a new branch out of main, and copy visualization/circuit/utils.py and visualization/circuit/matplotlib.py into the branch. Things to test,

• Try lots of complex existing circuits to make sure 1. above is satisfied.
• Create lots of if_test examples with and without else and using different kinds and complexities of circuits.
• Set fold=-1 to the draw command initially to test 2. This will do no folding.
• Set fold to a small number and see that the circuits properly wrap.

To do

• Tests?

Futures

• Add switch, for, and while control flow ops.
• Perhaps use some of this capability for showing sub-circuits or expanding gates built from circuits.

qr = QuantumRegister(4, "q")
cr = ClassicalRegister(3, "cr")
qc = QuantumCircuit(qr, cr)
qc.h(0)

with qc.if_test((cr[1], 1)) as _else:
    qc.x(0, label="X1111i").c_if(cr, 4)
    with qc.if_test((cr[2], 1)):
        qc.z(0)
        with qc.if_test((cr[1], 1)):
            qc.cx(0, 1)
            with qc.if_test((cr[2], 1)):
                qc.y(1)
                qc.z(2)
                with qc.if_test((cr[1], 1)):
                    qc.cx(0, 1)
with _else:
    qc.y(1)
    with qc.if_test((cr[2], 1)):
        qc.x(0)
        qc.x(1)
    qr1 = QuantumRegister(2, "qr1")
    cr1 = ClassicalRegister(2, "cr1")
    cr2 = ClassicalRegister(2, "cr2")
    circuit = QuantumCircuit(qr1, cr1, cr2)
    inst = QuantumCircuit(2, 2, name="Inst").to_instruction()
    qc.append(inst, [qr[0], qr[1]], [cr[0], cr[1]])

qc.x(0)
qc.draw(fold=30, style={"showindex": True})

Using fold=7, the above circuit looks like this,

Directly instantiating the IfElseOp instead of using if_test also works.

qr = QuantumRegister(4, "q")
cr = ClassicalRegister(3, "cr")
qc = QuantumCircuit(qr, cr)
qc.h(0)
qc.h(1)
qc.measure(0, 1)
qc.measure(1, 2)
qc.x(2)
qc.x(2, label="XLabel").c_if(cr, 2)

qr2 = QuantumRegister(3, "qr2")
qc2 = QuantumCircuit(qr2, cr)
qc2.x(1)
qc2.y(1)
qc2.z(0)
qc2.x(0, label="X1i").c_if(cr, 4)

qc.if_else((cr[1], 1), qc2, None, [0, 1, 2], [0, 1, 2])
qc.x(0, label="X1i")
qc.draw(style={"showindex": True}, cregbundle=True)

[qiskit-bot]

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

• @enavarro51
• @Qiskit/terra-core

[coveralls]

Pull Request Test Coverage Report for Build 5258853096

• 155 of 318 (48.74%) changed or added relevant lines in 2 files are covered.
• 20 unchanged lines in 6 files lost coverage.
• Overall coverage decreased (-0.04%) to 85.865%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
qiskit/visualization/circuit/matplotlib.py	153	316	48.42%

Files with Coverage Reduction	New Missed Lines	%
crates/qasm2/src/lex.rs	1	91.65%
qiskit/extensions/quantum_initializer/squ.py	2	80.0%
qiskit/transpiler/passes/synthesis/unitary_synthesis.py	2	90.18%
crates/accelerate/src/sabre_swap/layer.rs	4	97.32%
qiskit/visualization/circuit/matplotlib.py	5	52.01%
crates/qasm2/src/parse.rs	6	97.11%

Totals
Change from base Build 5256241220:	-0.04%
Covered Lines:	71470
Relevant Lines:	83235

---

💛 - Coveralls

[jakelishman]

Edwin and I spoke privately, and currently our intention is skip review of this PR and merge #10207 directly, which includes this PR and additionally reworks a fair amount of its internals.

I'll leave this PR unclosed for now, but our current intention is to close it via #10207.