Fix/benchmark creation tests

!222 (merged) introduced a new example in the docs section. but did not add it to the benchmarks (meaning test suite on develop now fails). Also the example used code blocks instead of jupyter-execute blocks, meaning the code in the example is never ran.

This MR adds some code to skip the inference example while creating the benchmarks, since it runs very slowly (I haven't actually ran it to completion yet)

It also adds a test that checks if all the created benchmark functions are actually being imported in the benchmark test module (before the script could run and generate the code to benchmark, but no check that it was actually being benchmarked)