How viable is quantum annealing for solving linear algebra problems?