Mulilabel limitation should be documented

[tralfamadude]

Only 7 labels are supported and this is not documented. Since effort can be expended to prepare training data, finding out this limitation when running train.py is wasteful.

[tralfamadude]

This is misnamed due to a misunderstanding: should be "Classification limitation should be documented".

More detail: You can have up to 7 labels. If you define more than that, an error in assigning values to the model will make training fail.