RDF generates an artificial peak at approx 1

[Alyssa Travitz]

With some systems there is a large peak at about r=1. I've attached a minimal script and a representative .pos file that replicates this error.

[Bradley Dice]

I am taking a look at this.

[Bradley Dice]

This is an issue in the original simulation data, not a problem with freud. A tabulated potential in HOOMD needs more data near small values of r, and it is artificially creating bound systems. (This is not a problem with HOOMD, just improper user input.)