Fix IndexError in apply_png_predictor for wide scanlines

[SAY-5]

Pull request

Fixes #1269. apply_png_predictor seeds the prior row with line_above = bytearray(columns), but a scanline is nbytes = colors * columns * bitspercomponent // 8 bytes wide. When that is wider than columns (multiple colors, or 8-bit components), the first scanline with filter type 3 (Average) or 4 (Paeth) indexes line_above[j] past its end and raises IndexError: list index out of range. Seeding it with bytearray(nbytes) gives the spec-mandated full-width zero prior row, matching the width of the raw rows used for every later scanline.

How Has This Been Tested?

Added test_apply_png_predictor_multibyte_pixels in tests/test_utils.py, which decodes a 3-color scanline (6 bytes wide, columns=2) with the Average filter and checks the output. It raises the reported IndexError without the fix and passes with it. The full tests/test_utils.py suite passes and ruff is clean.

Checklist

• I have read CONTRIBUTING.md.
• I have added a concise human-readable description of the change to CHANGELOG.md.
• I have tested that this fix is effective or that this feature works.
• I have added docstrings to newly created methods and classes.
• I have updated the README.md and the readthedocs documentation. Or verified that this is not necessary.

[nikolas]

This is a duplicate of #1271

[SAY-5]

You're right, #1271 has the same fix already in flight. Closing this one in favor of that.