CLN: Enhance pybind setup & move some more functions to C++

[mferrera]

No particular fixation with the bulkvol function, just have to start somewhere and this was a good spot. But the main focus of this PR is to expand the pybind11 setup into something more scalable. This format will probably change later, too. Minor performance improvement.

Resolves #1137 Adds to #1138

Old:

• On a 10 x 50 x 5 grid

  ------------------------------------------------- benchmark 'bulkvol': 1 tests ------------------------------------------------
  Name (time in us)               Min       Max      Mean   StdDev    Median      IQR  Outliers  OPS (Kops/s)  Rounds  Iterations
  -------------------------------------------------------------------------------------------------------------------------------
  test_benchmark_bulkvol     638.6250  819.0420  675.9939  26.0284  673.3340  48.2493     360;2        1.4793     843           1
  -------------------------------------------------------------------------------------------------------------------------------

• On a 100 x 500 x 25 grid

  --------------------------------------------- benchmark 'bulkvol': 1 tests --------------------------------------------
  Name (time in ms)               Min       Max      Mean  StdDev    Median     IQR  Outliers     OPS  Rounds  Iterations
  -----------------------------------------------------------------------------------------------------------------------
  test_benchmark_bulkvol     299.5485  301.1779  300.4839  0.7149  300.5362  1.2737       1;0  3.3280       5           1
  -----------------------------------------------------------------------------------------------------------------------

New:

• On a 10 x 50 x 5 grid

  ------------------------------------------------- benchmark 'bulkvol': 1 tests ------------------------------------------------
  Name (time in us)               Min       Max      Mean   StdDev    Median      IQR  Outliers  OPS (Kops/s)  Rounds  Iterations
  -------------------------------------------------------------------------------------------------------------------------------
  test_benchmark_bulkvol     500.6250  619.4590  515.5544  17.4468  507.0000  22.3338    124;21        1.9397     935           1

• On a 100 x 500 x 25 grid

  --------------------------------------------- benchmark 'bulkvol': 1 tests --------------------------------------------
  Name (time in ms)               Min       Max      Mean  StdDev    Median     IQR  Outliers     OPS  Rounds  Iterations
  -----------------------------------------------------------------------------------------------------------------------
  test_benchmark_bulkvol     239.4543  240.9973  240.2568  0.5707  240.3211  0.7316       2;0  4.1622       5           1
  -----------------------------------------------------------------------------------------------------------------------

Taking the medians at face value, in sum these changes made the function 30% faster from where it started with the SWIG C binding.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Comparison is base (0af8d2f) 80.28% compared to head (16df165) 80.30%.

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #1139 +/- ## ========================================== + Coverage 80.28% 80.30% +0.02% ========================================== Files 97 97 Lines 13571 13560 -11 Branches 2177 2174 -3 ========================================== - Hits 10895 10890 -5 + Misses 1954 1950 -4 + Partials 722 720 -2 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.