fix pytket 1.35 issues

[cqc-melf]

Description

Generating the QIR with pytket 1.35. gives for some circuits differences compared to the previous pytket version.

Related issues

Please mention any github issues addressed by this PR.

Checklist

• I have performed a self-review of my code.
• I have commented hard-to-understand parts of my code.
• I have made corresponding changes to the public API documentation.
• I have added tests that prove my fix is effective or that my feature works.
• I have updated the changelog with any user-facing changes.

[cqc-alec]

Do you know why these changes are happening?

[cqc-melf]

    c = Circuit(1, name="test_classical")
    a = c.add_c_register("a", 8)
    b = c.add_c_register("b", 10)
    d = c.add_c_register("d", 10)

    c.add_c_setbits([True], [a[0]])
    c.add_c_setbits([False, True] + [False] * 6, list(a))
    c.add_c_setbits([True, True] + [False] * 8, list(b))

    c.add_c_setreg(23, a)
    c.add_c_copyreg(a, b)

    c.add_classicalexpbox_register(a + b, d)  # type: ignore
    c.add_classicalexpbox_register(a - b, d)  # type: ignore
    # c.add_classicalexpbox_register(a * b // d, d)
    c.add_classicalexpbox_register(a << 1, a)  # type: ignore
    c.add_classicalexpbox_register(a >> 1, b)  # type: ignore

    c.X(0, condition=reg_eq(a ^ b, 1))
    c.X(0, condition=(a[0] ^ b[0]))
    c.X(0, condition=reg_eq(a & b, 1))
    c.X(0, condition=reg_eq(a | b, 1))

    c.X(0, condition=a[0])
    c.X(0, condition=reg_neq(a, 1))
    c.X(0, condition=if_not_bit(a[0]))
    c.X(0, condition=reg_gt(a, 1))
    c.X(0, condition=reg_lt(a, 1))
    c.X(0, condition=reg_geq(a, 1))
    c.X(0, condition=reg_leq(a, 1))
    c.Phase(0, condition=a[0])

[cqc-melf]

Above is the testcase for this QIR files, from my understanding the quantum gates get reordered, but I don't know why. From my understanding this is a generating an equivalent circuit.

[cqc-alec]

Oh I see, yes, reordering I can understand because of the change in classical expression handling.