Parametric weight function for DPD....

[bindgens1]

... with a parabolic weight function

Description of changes:

• New optional parameter "k" in DPDInteraction for the exponent for the weight function

[codecov]

Codecov Report

Merging #3570 into python will decrease coverage by <1%.
The diff coverage is 100%.

@@          Coverage Diff           @@
##           python   #3570   +/-   ##
======================================
- Coverage      88%     88%   -1%     
======================================
  Files         524     524           
  Lines       23598   23598           
======================================
- Hits        20772   20770    -2     
- Misses       2826    2828    +2

Impacted Files	Coverage Δ
src/core/dpd.hpp	100% <ø> (ø)	⬆️
src/core/dpd.cpp	91% <100%> (ø)	⬆️
src/core/electrostatics_magnetostatics/p3m.cpp	85% <0%> (-1%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update a4596e3...56362fc. Read the comment docs.

[fweik]

In principle the type parameter is obsolete now, no? type = 0 is just the same as k = 0.
Ah sorry, this is of course nonsens.

[bindgens1]

yeah 1- x^0 is zero and the weight function is totally gone.

Couldn't find a way to simplify it even more.

[fweik]

Ok, this looks good to me, but you should add a bit more documentation. In the main docs of the
DPD interaction (6.1.14. DPD interaction in doc/sphinx/inter_non-bonded.rst) you should
include the exponent factor into the formulas for the weight function and maybe add a literature
reference to the paper that describes the physical effect of changing the parameter.

[fweik]

LGTM, thank you for your contribution. I'll take it from here.