Optimize Morton order computation with hypercube optimization, vectorized decoding, and singleton dimension removal, providing 10-45x speedup for typical chunk shapes.