TicTacToe winning confition fix & ShowBoard function readability improvement

[ubiieta]

This pull request includes two main improvements to the TicTacToe reinforcement learning implementation:

1. Fix to winner() function:

   • Corrects a bug where diagonal wins (both directions) were not properly detected for the human player.
   • Now the agent correctly announces a win when the human player completes a diagonal line.

2. Update to showBoard() function:

   • Board cells now display a unique number (1–9) instead of requiring both row and column input.
   • Simplifies gameplay: users now input a single number to choose a move.