Minimizing the sum of absolute lateness in single‐machine and multimachine scheduling