[python] Calling builder kernels within decorator kernels is not supported

Required prerequisites

• Consult the security policy. If reporting a security vulnerability, do not report the bug using this form. Use the process described in the policy to report the issue.
• Make sure you've read the documentation. Your issue may be addressed there.
• Search the issue tracker to verify that this hasn't already been reported. +1 or comment there if it has.
• If possible, make a PR with a failing test to give us a starting point to work on!

Describe the bug

The following tests fail with the error "error: unhandled function call - builder_kernel".

def test_call_builder_kernel_from_decorator_kernel():
    """Tests that a kernel built with make_kernel can be called
    from within a @cudaq.kernel decorated function."""
    builder_kernel, qubit = cudaq.make_kernel(cudaq.qubit)
    builder_kernel.x(qubit)

    @cudaq.kernel
    def caller():
        q = cudaq.qubit()
        builder_kernel(q)

    counts = cudaq.sample(caller, shots_count=10)
    assert counts["1"] == 10


def test_call_builder_kernel_from_decorator_kernel_2():
    """Tests that a parameterized builder kernel can be called
    from within a @cudaq.kernel decorated function."""
    builder_kernel, angle = cudaq.make_kernel(float)
    qubits = builder_kernel.qalloc(1)
    builder_kernel.ry(angle, qubits[0])

    @cudaq.kernel
    def caller():
        builder_kernel(np.pi)

    counts = cudaq.sample(caller, shots_count=10)
    assert counts["1"] == 10

Steps to reproduce the bug

run the above test

Expected behavior

tests pass, or a better error is raised

Is this a regression? If it is, put the last known working version (or commit) here.

Not a regression

Environment

• CUDA-Q version:
• Python version:
• C++ compiler:
• Operating system:

Suggestions

No response

Start by running the two reproducer tests, test_call_builder_kernel_from_decorator_kernel and test_call_builder_kernel_from_decorator_kernel_2, to confirm the unhandled function call error. Trace how builder kernels created with cudaq.make_kernel are invoked from @cudaq.kernel functions; done means both tests pass or a clearer error is raised.