Search for both basic and bottleneck blocks (to fix "no known network structure detected" warning with ResNet-50 and other similar models)

[Lissanro]

Currently when trying to use ResNet-50 (the same issue happens with usual
ResNeSt-50 and many other similar models) from timm library
https://github.com/rwightman/pytorch-image-models, I get the following warning:

"Manifold mixup: no known network structure detected, 126 modules will be used
for mixup"

This happens because ResNet-50 and higher do not have any BasicBlock and have
Bottlenecks instead. With this patch I get this message:

"Manifold mixup: Block structure detected, 16 modules will be used for mixup."

I get better error_rate than when 126 modules were used for mixup.

[nestordemeure]

I would prefer to not need regexp for that (mostly to improve readibility and reduce the number of dependencies that are loaded). Do you have the time to implement it with basic strings operations only (the is_block code can be used as an example, it should be fairly fast to write) ?

If you do not, I will merge it in its current state.

[Lissanro]

Sure, no problem. Done. I also updated documentation to mention new behavior.

[nestordemeure]

Perfect!