Maxima Function
flush (expr, tensor_1, tensor_2, ...)
Set to zero, in expr, all occurrences of the tensor_i that have no derivative indices.