Improve performance

[Gram21]

Right now, the performance for bigger models is getting worse and the tool scales rather bad. This needs to be improved.

My current (educated) guess is that certain elements (individuals, classes, properties) are queried by name each time instead of using the URI/IRI. This was already detected in another project to be rather imperformant when the ontology gets bigger. Here, querying using the URI/IRI should be preferred.

Moreover, one can think about other methods like parallelizing some parts. However, this needs changes to the current algorithm.

Lastly, on the long run one can exchange the local ontology connector with https://github.com/InFormALin/Ontology-Connector. There are some performance tweaks and other improvements