jpeg8_encode fails when specifying bitspersample

[erikogabrielsson]

Hi,

With 2024.09.22 this simple use of jpeg8_encode fails:

from imagecodecs import jpeg8_encode
import numpy as np

data = np.zeros((128, 128), dtype=np.uint8)
jpeg8_encode(data, bitspersample=8, lossless=False)

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Cell In[15], [line 5](vscode-notebook-cell:?execution_count=15&line=5)
      [2](vscode-notebook-cell:?execution_count=15&line=2) import numpy as np
      [4](vscode-notebook-cell:?execution_count=15&line=4) data = np.zeros((128, 128), dtype=np.uint8)
----> [5](vscode-notebook-cell:?execution_count=15&line=5) jpeg8_encode(data, bitspersample=8)

File imagecodecs\\_jpeg8.pyx:161, in imagecodecs._jpeg8.jpeg8_encode()

ValueError: invalid bitspersample=8

Maybe this check is not as intended?
https://github.com/cgohlke/imagecodecs/blob/master/imagecodecs/_jpeg8.pyx#L157

[cgohlke]

Thanks for reporting! That case should of course work. It looks like a regression introduced in #105. The fix is going to be in the next release.