There's a big issue that all old popular image formats are 8-bits. 10-bits or even 12-bits would help a lot with storing more information and maintaining editability.
If adding more bits to an image format, please make it 'n bit'. Ie. the file could be 8 bit, it could be 10, it could be 12, it could be 60 bit!
Whilst we're at it, please get rid of RGB and make it N channels too.
Libraries can choose to render that into a 3 channel, 8 bit buffer for legacy applications - but the data will be there for CMYK or HDR, or depth maps, or transparency, or focus stacking, or any other future feature!
