Split project in LLM-VM and LLM-VM-server

To make it simple to use it would be nice to have the core working with LLM-VM objects and have a server project that uses the core models. Each project should have only the required dependencies. This would also fix the issue #152, given the main issue with installing it now on Google Colab is a flask dependency.

Methods from LLM-VM core should not return data in json / dict format but proper objects, which will make it easier to use LLM-VM as an external library.