CUDA-Q MPI Submodule yields incorrect results - the finalize command does not work

[zohimchandani]

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

import cudaq 

print('should yield false', cudaq.mpi.is_initialized())

cudaq.mpi.initialize()

print('should yield true', cudaq.mpi.is_initialized())

cudaq.mpi.finalize()

print('should yield false', cudaq.mpi.is_initialized())

should yield false False
should yield true True
should yield false True

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

[zohimchandani]

Can we also please include mpi.barrier commands in CUDA-Q to ensure we don't have the script jump ahead when we don't want it to.

[1tnguyen]

Re: cudaq.mpi.is_initialized() yields True after cudaq.mpi.finalize()

This is an expected behavior of MPI. cudaq.mpi.is_initialized is a wrapper of MPI_Initialized function.
According to the MPI spec:

MPI_Initialized: This routine may be used to determine whether MPI_INIT or MPI_INIT_THREAD has
been called. MPI_INITIALIZED returns true if the calling process has called either of these
MPI procedures. Whether MPI_FINALIZE has been called does not affect the behavior of
MPI_INITIALIZED.

We can expose the is_finalized API to Python to detect whether finalize has been called. This API is available in CUDA-Q C++ but not exposed to Python yet.