Moptipy

improved system information

- We added function for checking whether we are in a make build.
This allows adding dependencies only if we are not in a make build.
This is good, because during a make build, our package may not be installed.
- We now also store the command line in the log files.
This makes it easier to understand how the experiment was executed.

several minor improvements and fixes

In this release, we put the first steps into implementing surrogate model based continuous optimization methods in form of our [RBF interpolation](https://thomasweise.github.io/moptipy/moptipy.algorithms.so.vector.surrogate.html#module-moptipy.algorithms.so.vector.surrogate.rbf_interpolation) method.
The idea is to use [SciPy's RBF interpolation](https://docs.scipy.org/doc/scipy/reference/generated/scipy.interpolate.RBFInterpolator.html#scipy.interpolate.RBFInterpolator) as model.
The first draft algorithm (see [here](https://thomasweise.github.io/moptipy/moptipy.algorithms.so.vector.surrogate.html#module-moptipy.algorithms.so.vector.surrogate.rbf_interpolation)) then begins by sampling a few points and evaluating them with the original objective function.
All points that are evaluated by the original objective function as well as their objective values are always recorded.
We then create an RBF interpolation model based on these points.
We then apply the optimization algorithm using the model as objective function.
The best point that was sampled (according to that model objective function) is then evaluated again using the original objective function (and recorded).
Together with the other points evaluated by the original objective function, it is used to build a new model.
This process is repeated until the termination criterion is met.

numba uses intel-cmplr-lib-rt to generate faster compiled code, see https://numba.readthedocs.io/en/stable/user/performance-tips.html.
So we require it now as dependency as well.

bugfix for the logging of the H table in the (1+1) FEA: offset lower bound now added

now also passing instance to on_completion