Gets a bound on the optimal solution value of the active problem. When a model has been solved to optimality, this value matches the optimal solution value. Before optimality has been proven, this value is computed for a minimization (maximization) problem as the minimum (maximum) objective function value of all remaining unexplored nodes.