Move from pybind11 to nanobind

[Joseph-Edwards]

In the past few years, the author of pybind11 has been working on a new project: nanobind. From the documentation:

nanobind is a small binding library that exposes C++ types in Python and vice versa. It is reminiscent of Boost.Python and pybind11 and uses near-identical syntax. In contrast to these existing tools, nanobind is more efficient: bindings compile in a shorter amount of time, produce smaller binaries, and have better runtime performance.

The main difference [between pybind11 and nanobind] is a change in philosophy: pybind11 must deal with all of C++ to bind legacy codebases, while nanobind targets a smaller C++ subset. The codebase has to adapt to the binding tool and not the other way around, which allows nanobind to be simpler and faster.

As the features provided by libsemigroups_pybind11 grows, so too does the time taken to compile. On my machine, it currently takes around two minutes. Since nanobind has an almost identical interface to pybind11, we could potentially explore it as an option if the current build time gets too large.

I'm opening this discussion to make note of this idea, and invite other thoughts. This is NOT something that I think is something we should focus our attention on right now, rather a thought to keep in the back of our minds.