"lim sup", "sup", "lim inf", "inf"

[Happypig375]

Unless there is something before 3.1 that specifically requires these, let's consider this in v3 design but start the concrete work later

[Rafael-SOWNet]

Nothing before 3.1 requires them. Neither the integrator's work nor Sullivan and Mackey's book (#1409) uses a supremum, infimum, lim sup or lim inf. The book's one exercise that touches the idea, a union of intervals [1, (i + 1)/i], is already answered by the rule for families of intervals with monotone ends. The closest thing the library has is max and min of an expression over a set (#1212).

[Happypig375]

Related to #1534 - these are operations on infinities. I guess I'll spawn a 3.2 milestone for now

[Happypig375]

Ok, I spawned the 3.2 milestone (and edited 2.7 such that indefinite integration is to be done in 2.6). Post-3.1 releases should have a theme each release for effective communication on the concrete area of improvement.