Fioletov approach, big one

Created by: enricodammers

I merged all my local fioletov approach changes into a single package. I adapted Kevin's style as much as possible, and the output of the approach should fit into your web tool. I've pulled the code through several of VSCs checkups for proper spacings etc. No christmas tree on my side.

You'll probably want to move around some parts of tools.py to other scripts, should be easy enough to fix. I've had some issues with the local set of TROPOMI observations, so i kept the create_subset tests commented out in test_tools.py.

I'm thinking about how to properly add the error analysis. the clean way to do it is running a monte carlo like approach for the various parameters but thats quite expensive on this server. Tests in the past showed that lifetime is dominant so we could potentially approximate fro now. A few TODO!'s here and there in the code. lets walk through it during the next meeting and see.

Ignore the .git file that sneaked in, forgot to remove it from my commit list.

Anyway take a look, and speak to you all tomorrow.