Use IntervalSets.jl for set operations

[dpsanders]

cf. discussion in JuliaMath/IntervalSets.jl#55

This is related to redefining .. to not do clever tricks and just take floating-point endpoints at face value.

[lbenet]

For reference, see the discussion in #345.

Regarding using IntervalSets.jl for set operations, would that satisfy the standard?

[Kolaru]

For most of it, it should, but some operation would need to be redefined. For example, for us, union of disjoint interval should return the hull instead of erroring.

While I like the idea of building intervals on a more fundamental package, I feel like in the current state we would loose too much flexibility for a rather small gain. More precisely, given the amount of manipulation we perform of the interval constructor and the restriction we impose on the bounds, I think that type piracy is not an option. And without type piracy, we can not share .. and would need to wrap IntervalSets intervals in ours to use the method defined for them.

[Kolaru]

Discussed in today's call, in favor of status quo.