Add Lennard-Jones parameters to atomtypes

[timholy]

This extracts more data from the Amber force field.

Also fixes a path inconsistency in the data extraction.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 95.19%. Comparing base (8802eb7) to head (caec19f).

Additional details and impacted files

@@           Coverage Diff           @@
##           master      #91   +/-   ##
=======================================
  Coverage   95.19%   95.19%           
=======================================
  Files          15       15           
  Lines        2185     2185           
=======================================
  Hits         2080     2080           
  Misses        105      105           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[jgreener64]

I do have reservations about reading this data, as it is force field specific rather than a topological property of the system. There is a grey area, as discussed in #73, since for example we read bond distances. However, these tend to be very similar between force fields and have a spatial interpretation, whereas the Lennard-Jones sigma and epsilon are parameters of the force field (though sigma does have a link to the vdw radius).

If we read these properties we should probably read in all the force field parameters, which is more the domain of Molly.jl. The MolecularForceField type there will read in everything, a quite involved process, but that is a much heavier dependency than BioStructures.jl.

[timholy]

That's a fair concern; I'd forgotten about the slippery slope. Distinguishing topology and forces seems like the right line to draw.

Closing.