v1.2.0

Firstly the game has been modified to adjust the neural network weightings, if the human player wins (playing in hardest mode). The weighting changes make the computer play slightly differently, so a player can't keep repeating the same set of winning moves.

As the changes make the game more difficult, easy, medium and hard options have been added. Easy and medium use occasional random moves (i.e. mistakes) to allow to give the player a chance.