[python] Raise a clear error from to_cupy when CuPy is missing

[udsy19]

Summary

When CuPy is not installed, cudaq.to_cupy raises a confusing NameError about cp instead of the intended "please install CuPy" message.

Root cause

try:
    import cupy as cp
except ImportError:
    print('to_cupy not supported, CuPy not available. Please install CuPy.')

if dtype is None:
    ...
    dtype = cp.complex128 if ... else cp.complex64   # cp was never bound -> NameError

The except branch prints and then falls through, so the next use of cp raises NameError: name 'cp' is not defined, masking the real cause.

Fix

Raise a RuntimeError with the existing message (matching the RuntimeError already raised later in the same function for the not-on-GPU case), so the import guard fails clearly before anything else is touched.

Tests

Added python/tests/builder/test_to_cupy_missing_cupy.py, which forces the cupy import to fail (via __import__ monkeypatch) regardless of whether CuPy/GPU are present, and asserts a clear RuntimeError mentioning CuPy.

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.