Dimension 1 bug

[nilgoyette]

I'm not entirely happy with this fix because I don't understand why it works :| It make sense that writing the data bytes of a (1, 5, 5, 5) image is the same as writing them for a (5, 5, 5) image. The first dimension adds nothing interesting that should concern the write_data function. However, it looks like iterating on a dim=1 change the ordering. When I skip the dim=1, it works as intended.

Anyway, I'm still pushing this branch because I have other things to do, but if you have a better understanding of the problem than me, I'm interested!

Fix #63

[nilgoyette]

I requested you as reviewer because I don't even know if you saw this PR.

[Enet4]

Sorry for the delay. I had looked into this issue, and indeed reproduced it on my machine, but in my attempts to better understand what was going on, it got postponed almost indefinitely.

Considering that the crate is better off with this tweak than without any solution, I agree to merge. We can keep on the look out for similar issues, chances of something else being at play might not be entirely nil.

Thank you very much, and sorry for the time inconvenience.