print_graph method displays the Smart Answer as a tree.
print_questions_and_possible_next_nodes method displays the questions and their possible next nodes.
The extraction of possible next nodes is horrible in the case where the Smart Answer uses the block syntax of
next_node. I'm getting the source of the block and extracting things that look like symbols.