Allow generators into Model's fit, evaluate and predict

[lucasdavid]

Summary

I've ported the interesting sections of the code implemented in tensorflow/keras/engine/training.py#L749
and some necessary utils. Generators and Sequences can now be passed to Model#fit, Model#evaluate and Model#predict.

The parameters max_queue_size, workers, use_multiprocessing were added to these methods (they now have the exact same signature as the ones in tf.keras.

Related Issues

#11772

PR Overview

• [y] This PR requires new unit tests [y/n] (make sure tests are included)
• [n] This PR requires to update the documentation [y/n] (make sure the docs are up-to-date)
• [y] This PR is backwards compatible [y/n]
• [n] This PR changes the current API [y/n] (all API changes need to be approved by fchollet)

[fchollet]

Thanks for the PR. When you copy code from tf.keras, please make sure to adapt it: for instance, tf.data references are not relevant, docstring formatting is different, etc.

[lucasdavid]

@fchollet everything's fixed now.

[fchollet]

Excellent, thank you!