The magic constants used in the bit-twiddling example also have a very regular derivation. Designing an algorithm that computes the correct constants for Morton Numbers of arbitrary dimensionality is pretty simple.
There is a neat generalization of this algorithm that extends it to irregular bit interleaving patterns at the cost of requiring a few more magic constants (also derivable). I once wrote a compact engine that used algorithmically generated transform constants to produce arbitrary bit interleaving patterns in an arbitrary number of dimensions via these bit-twiddling algorithms.
There were some recent submissions here about related topics like space-filling curves and secret sharing that I've looked at before. When I started looking into abstract algebra and came across finite fields I sort of laughed at the Wikipedia page mentioning right at the beginning "Finite fields are important in number theory, algebraic geometry, Galois theory, cryptography, coding theory and Quantum error correction" with each topic being linked. It's a fun way to procrastinate.
If your engine isn't part of some secret sauce I think it would be neat to study it. Unless you know of an already existing open source equivalent that's not hidden away in the corners of something like GCC?
Once you know that, you can generate any distribution you need. However, Morton numbers take advantage of a property of their bit distributions where intermediate steps never corrupt a bit that is not going to be masked off anyway. This is not true for some other bit interleaving patterns. Extending it to arbitrary patterns requires two masking operations with another magic constant at each step which protects bits that would otherwise be destroyed in the simple Morton algorithm.
Nothing like the transform engine I wrote is open source to the best of my knowledge. There is no reason I could not open source it, I just haven't. It is pretty efficient in that it can do a number of reductions and simplifications to the minimum number of steps and constants required to produce an n-dimensional interleaved result. For irregular patterns, you end up with quite a few no-op steps that can be eliminated; for regular patterns you can reuse steps, saving memory. I'll probably write it up and put it on the web at some point.