Implement analyze FEP verbose to investigate convergence

[David-Araripe]

Also recommended in the best practices, see: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8388617/#:~:text=Convergence%20of%20simulations

Some adjustments might be needed since we have a different sampling technique as compared to most other people working with FEP. Some input from Willem would be appreciated, but let's keep this on the to-do list...

[jesperswillem]

We usually look at the sem, I have also been toying around at some point with plotting the energetic profiles (somewhat similar to the lambda window overlap). Happy to discuss, the code to generate plots is in analyze_FEP.py