Fix plot_rgb stretch failure with masked NaN arrays

[xiangcheng1112]

IMPORTANT: DO NOT USE THIS REPO TO PRACTICE PULL REQUESTS AND LEARN GITHUB!

Earth Lab -- Pull Request Template

Welcome to EarthPy!

Before submitting a Pull Request please be sure to submit an issue that describes what the contents of the PR addresses.

IMPORTANT: Please do not use this repository as a practice activity to learn how to use GitHub! Only submit a PR after:

1. Submitting an issue with a valid item that you wish to address in earthpy
2. Getting approval from one of the maintainers to submit a formal PR.

Description

Fixes #675

_stretch_im() uses np.nanpercentile() which does not handle masked arrays correctly.
When a band is a masked array (e.g., raster loaded by rioxarray), it returns NaN
values, causing rescale_intensity to produce a black image.

Added a check: if band is masked, use np.percentile(band.compressed(), ...) instead.

Fixes # (issue)

Type of change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)

How Has This Been Tested?

Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration

Testing

Test A - Existing test suite

python -m pytest earthpy/tests/test_plot_rgb.py -v
All 12 tests passed, including test_stretch_image_nan and test_masked_im.
No regressions.

Test B - Reproduction of issue #675

import matplotlib
matplotlib.use('Agg')
import earthpy.plot as ep
import numpy as np

r = [[128,  64, np.nan], [128,   0,   0], [  0, 128, 128]]
g = [[128,  64, np.nan], [  0, 128,   0], [128,   0, 128]]
b = [[128,  64, np.nan], [  0,   0, 128], [128, 128,   0]]
arr = np.array([r, g, b])

arr_nan_masked = np.ma.masked_invalid(arr)
ax = ep.plot_rgb(arr_nan_masked, rgb=(2, 1, 0), stretch=True)
img = ax.get_images()[0].get_array()
assert img.max() > 10, "Image should not be all black"
print("PASSED: image min=%.1f max=%.1f" % (img.min(), img.max()))
Previously returned a black image. Now renders correctly (min=0.0, max=255.0).

Test configuration

- OS: Windows 11
- Python: 3.11.9
- Key dependencies: numpy, matplotlib, scikit-image, rasterio


## Checklist:

- [ ] I have already submitted an issue and it was approved for a pr by an earthpy maintainer
- [x ] My code follows the style guidelines of this project
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings
- [ ] I have added tests that prove my fix is effective or that my feature works
- [x] New and existing unit tests pass locally with my changes
- [ ] Any dependent changes have been merged and published in downstream modules
- [x ] I have checked my code and corrected any misspellings