Lagrangian Penalization Scheme with Parallel Forward–Backward Splitting