You are given a puzzle consisting of a row of squares that contain non-negative integers, with a zero in the rightmost square. You have a token that starts on the leftmost square. On each turn, the token can shift left or right a number of squares equal to the value in its current square, but is not allowed to move off either end. For example, if the row of squares contains these values: [2, 4, 5, 3, 1, 3, 1, 4, 0], then on the first turn the only legal move is to shift right two squares, because the starting square contains a 2, and the token can't move off the left end. The goal is to get the token to the rightmost square (that contains zero). This row has a solution (more than one), but not all rows do. If we start with the row [1, 3, 2, 1, 3, 4, 0], then there is no way for the token to reach the zero. Write a recursive function named row_puzzle that takes a list of integers as a parameter and returns True if the puzzle is solvable for that row, but returns False otherwise.