Add oxygen evolution utility function

[esoteric-ephemera]

Adds a utility to recompute oxygen evolution according the method used previously in emmet

A couple of notes about this implementation of get_oxygen_evolution: this takes in an MPID, working ion, and optional thermo type (defaults to GGA / GGA+U mixed hull). On output, a dict with this structure is output:

{
  entry_formula : {
    "mu": [list of chemical potentials],
    "reaction": [list of pymatgen reaction objects, normalized to integer coefficients],
    "evolution" : [list of floats describing how much O1 has been evolved in the reaction],
    "O2_produced": [list of floats describing how much O2 has been produced]
  }
}

The negative of the "mu" list should probably be plotted against the "O2_produced" field. Not sure what's best here

Tagging @minhsueh and @tsmathis for comments on data organization / structure

[tsmathis]

I think the output looks fine. If this works well we can eventually move this into the insertion electrode document model to reduce the work being done server side