Optimal Game Move Strategy

In combinatorial game theory, adversarial search algorithms are essential for developing intelligent agents that compete against opponents. One of the most elegant approaches is the minimax strategy, which models decision-making in zero-sum games where one player's gain is exactly the other player's loss.

Consider a two-player, turn-based grid game played on a 3×3 board. Each player places their symbol ('X' or 'O') on empty cells, taking turns until one player achieves three consecutive symbols in a row, column, or diagonal, or all cells are filled (resulting in a draw).

The Adversarial Search Principle: The core idea behind optimal play is to assume that your opponent will always make the best possible move from their perspective. Given this assumption:

• The maximizing player aims to achieve the highest possible score (win)
• The minimizing player (opponent) aims to achieve the lowest possible score for the maximizer (blocking wins, forcing draws, or winning themselves)

Game State Evaluation: At each game state, the algorithm recursively explores all possible future moves, propagating values up the decision tree:

• Win for current player: +1 (or +∞)
• Loss for current player: -1 (or -∞)
• Draw: 0

Your Task: Implement an adversarial search function that determines the optimal next move for a given player on a 3×3 grid game board. The function should:

• Analyze all possible game continuations assuming perfect play from both sides
• Return the cell coordinates (row, column) that give the current player the best possible outcome
• Handle any valid game state where it's the specified player's turn

Note: If multiple moves are equally optimal, any one of them is an acceptable answer. Do not use external game-playing libraries.