feat: Implement Eq and Hash for Property

[RobertJacobsonCDC]

This is an experimental branch that is compatible with this ixa branch. It changes all instances of f64 to use OrderedFloat<f64>.

This PR only exists for visibility. We won't merge this.

---

It turns out that using OrderedFloat<f64> is really obnoxious. Even though OrderedFloat<f64> implements all the usual operations, you still can't mix f64 and OrderedFloat<f64> for comparisons, for example. OrderedFloat<f64> implements Deref<f64> / DerefMut<f64>, but that's not a silver bullet. You end up either calling into to upcast your f64 values to OrderedFloat<f64>, or you downcast your OrderedFloat<f64> values to f64. And both ways of downcasting are obnoxious in mathematical code! You have your choice of a deref operator, an *, which looks like multiplication, or using the tuple field accessor syntax, ordered_float.0, which just as gross when you use it in a mathematical formula. You could try to only ever use OrderedFloat<f64>, avoiding naked f64 entirely. But you don't always have that option. Both the standard library and ixa uses f64 for various things like time for example. The ordered-float crate does have a feature flag for rand support, though.

Local Benchmark Results

main (33ab8db) vs RobertJacobsonCDC_eq_hash

Benchmark	Base	Current	Change
init/population/10k	3.4157 ms	3.1341 ms	-8.2% faster ✓
init/population/100k	44.433 ms	41.496 ms	-6.6% faster ✓
execute/population/10k	81.324 ms	78.589 ms	-3.4% faster ✓
execute/population/100k	1.2189 s	1.1451 s	-6.1% faster ✓