Hamiltonian list does not work with observe_async

Required prerequisites

• 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 code snippet below calculates expectation values using observe and observe_async and executes fine

import cudaq
from cudaq import spin 

kernel = cudaq.make_kernel()

qubits = kernel.qalloc(2)

kernel.x(qubits[0])

kernel.y(qubits[1])

hamiltonian = spin.z(0)

result1 = cudaq.observe(kernel, hamiltonian).expectation_z()

result2 = cudaq.observe_async(kernel, hamiltonian, qpu_id = 0).get().expectation_z()

result1 == result2

If I extend the hamiltonian to a list of 2 elements, I can successfully calculate the expectation value as such:

import cudaq
from cudaq import spin 


kernel = cudaq.make_kernel()

qubits = kernel.qalloc(2)

kernel.x(qubits[0])

kernel.y(qubits[1])

hamiltonian = [spin.z(0), spin.z(1)]

result = cudaq.observe(kernel, hamiltonian)

exp_val = [result[i].expectation_z() for i in range(len(result))]

However, with obersve_async the following error is produced:

import cudaq
from cudaq import spin 


kernel = cudaq.make_kernel()

qubits = kernel.qalloc(2)

kernel.x(qubits[0])

kernel.y(qubits[1])

hamiltonian = [spin.z(0), spin.z(1)]

result = cudaq.observe_async(kernel, hamiltonian, qpu_id = 0)

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[2], line 15
     11 kernel.y(qubits[1])
     13 hamiltonian = [spin.z(0), spin.z(1)]
---> 15 result = cudaq.observe_async(kernel, hamiltonian, qpu_id = 0)

TypeError: observe_async(): incompatible function arguments. The following argument types are supported:
    1. (kernel: cudaq._pycudaq.Kernel, spin_operator: cudaq::spin_op, *args, qpu_id: int = 0, shots_count: int = -1, noise_model: Optional[cudaq::noise_model] = None) -> cudaq::async_result<cudaq::observe_result>

Invoked with: <cudaq._pycudaq.Kernel object at 0x7f0c50ff2770>, [<cudaq._pycudaq.SpinOperator object at 0x7f0c50fd1630>, <cudaq._pycudaq.SpinOperator object at 0x7f0c4c674e30>]; kwargs: qpu_id=0

observe_async supports type(spin.z(0)) which is cudaq._pycudaq.SpinOperator but the type of hamiltonian = [spin.z(0), spin.z(1)] is a list.

This is also reflected in the documentation where observe can accept a list of a spinop but observe_async only accepts spinop

Steps to reproduce the bug

NA

Expected behavior

NA

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

Not a regression

Environment

NA

Suggestions

No response

The issue identifies the Python observe_async API and its documentation as entry points; start by comparing its accepted spin_operator types with observe's list handling. Add regression coverage for a list of spin operators and verify that the asynchronous call returns an expectation value for each element, then ensure the documented behavior matches the supported interface.