Replace imghdr with something like filetype

[normanlorrain]

https://docs.python.org/3/library/imghdr.html (deprecated; good until Python 3.13)

filetype:
https://pypi.org/project/filetype/#supported-types

Rationale:
we need to determine the size of images.

[normanlorrain]

Current pytest suite shows where the error is.

[FlominatorTM]

I have at least one jpeg image, where elif imghdr.what(fname) returned None. Since there's a try-except clause with an empty return value anyway, it wouldn't hurt to add at least or imghdr.what(fname) == None: to that condition. I would assume that most images are jpeg, so it should cover most of the cases like mine.

mdpdf/mdpdf/image.py

Line 22 in a7a6ed4

elif imghdr.what(fname) == "jpeg":

On the other hand an implementation with guess = filetype.guess(fname) and then a switch using guess.extension might be more stable.

[normanlorrain]

I've replaced the code with a much simpler solution. Pytest suite passes. This is better anyway, using the already included PyMuPDF library, and the native imghdr module is deprecated.

Let me know if your jpeg still causes this code to stumble. Thanks!