Are Neural Rankers still Outperformed by Gradient Boosted Decision Trees?