martinate.sh doesn't include insane.py

[OLaprevote]

Assembling all the bricks for using martini forcefield on a protein feels like a convoluted treasure hunt, and having insane.py here would simplify it greatly.

There would be some incentives to having insane.py directly in this script:

• It isn't yet on GitHub, and needs some change work with martini3.
• martinate.sh requires it to run anyway, so anyone downloading this repository will HAVE TO search for insane.py, then put it in gromit repository, rename it insane and make it executable with a chmod +x insane. This felt annoying for me, but may have been hard for a beginner in informatics who would just want to do some coarse-grained simulations.
• The very least would be to give a link to where to download insane.py in the README.md, which currently isn't even the case, so you have to go on cgmartini.nl and find insane.py link . (For anyone reading this issue and still wondering where insane.py was hidden: in Downloads -> Tools -> Proteins and bilayers.)

I added insane and made changes in README in Pull Request #25. Still pending for review.

Another possibility would be to create its own personal GitHub repository to insane.py, then have it here as a git submodule.

[OLaprevote]

We need a Git repository for insane.py because it's deprecated for martini3 forcefield. Currently insane.py provided on ngmartini.nl:

1. Doesn't work on python3, so can't be used with martinize2 and the latest Martini3 forcefield. One easy fix is to run 2to3 -w insane.py. 2to3 is an executable that comes with python to easily pass python scripts from python 2 to python 3, and it does a remarkable job on insane.py (see Issue martinate.sh requires python 2 while martinize2 requires python 3 #28).
2. Won't produce usable gro and top files with Martini3 forcefield. Ions in Martini3 forcefield are named 'NA', 'CL', etc while in Martini2 they were named 'NA+', 'CL-', etc (see martini_v3.0.0_ions_v1.itp in martini_v300 archive, and martini_v2.0_ions.itp).

Worst thing is that I have literally no idea WHERE I should state this. On martini forum? Here because martinate.sh uses insane.py?

One good thing about adding insane.py directly in gromit would be that this comment wouldn't be just providing issues with current insane.py to stress the necessity of adding it on GitHub: it would also become a valid issue for gromit.