feat: GPU array processing with Cupy

[PinkShnack]

As done in qpretrieve, I would like to enable nrefocus to process 3D arrays, and have the option to process them on the GPU. Therefore, this will reflect qpretrieve's changes quick closely as described in #23.

To-do

• update changelog
• add docs describing cupy additions
• Add cupy backend with xp syntax
• make sure the expected backend attributes are present
• import protections for pyfftw and cupy
• simple ndarray backend swap for user, as done in qpretrieve
• Add 3D array processing in separate PR
  • Decide if autof.py needs changing

[codecov-commenter]

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

❌ Patch coverage is 67.54967% with 49 lines in your changes missing coverage. Please review.
✅ Project coverage is 79.96%. Comparing base (3422d02) to head (3c7d565).
⚠️ Report is 7 commits behind head on master.

Files with missing lines	Patch %	Lines
nrefocus/iface/rf_cupy.py	0.00%	25 Missing ⚠️
nrefocus/legacy.py	0.00%	12 Missing ⚠️
nrefocus/_ndarray_backend.py	71.87%	9 Missing ⚠️
nrefocus/iface/__init__.py	77.77%	2 Missing ⚠️
nrefocus/iface/rf_numpy_1d.py	85.71%	1 Missing ⚠️
❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@            Coverage Diff             @@
##           master      #24      +/-   ##
==========================================
- Coverage   84.46%   79.96%   -4.50%     
==========================================
  Files          21       22       +1     
  Lines         502      569      +67     
==========================================
+ Hits          424      455      +31     
- Misses         78      114      +36     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[PinkShnack]

@paulmueller I have added Cupy to nrefocus here. The 3D array processing I will do in another PR. Unfortunately the branch name is therefore a little off, but I feel like that's ok. Please review if you have the time. Very similar to the qpretrieve setup

[PinkShnack]

tests fresnel dataset created via:

def test_2d_refocus1_fresnel():
    pixel_size = 1e-6
    rf = nrefocus.RefocusNumpy(field=np.arange(256).reshape(16, 16),
                               wavelength=8.25*pixel_size,
                               pixel_size=pixel_size,
                               medium_index=1.533,
                               distance=0,
                               kernel="fresnel",
                               padding=False)

    refocused = rf.propagate(distance=2.13*pixel_size)
    refocused_flat = np.array(refocused).flatten().view(float)
    np.savetxt(data_path / "test_2d_refocus1_fresnel.txt", refocused_flat)