Updating charRNN pre-trained models

[shiffman]

The directory of pre-trained charRNN models are being used for a homework assignment in my Intro to ML for the Arts course. This pull requests makes the following changes:

• changes directory name from LSTM to charRNN to match ml5.js naming conventions.
• removes the Harry Potter pre-trained model (it is the only example of a modern fiction series and questions around copyright and permission to host such a model are unresolved.)
• A README file is added referencing important readings to consider when working with pre-trained charRNN models trained off of historical texts.

[shiffman]

I'm going to go ahead and merge this without review since I am assigning it for my course, but welcome feedback and changes in this thread or as separate issues / pull requests.