EVT for rV

[yosakaon]

Motivation for this change

This is a generalization of the Extreme Value Theorem already present for real-valued functions to R^n.
Should we keep both statements ?

Checklist

• added corresponding entries in CHANGELOG_UNRELEASED.md

• added corresponding documentation in the headers

Reference: How to document

Merge policy

As a rule of thumb:

• PRs with several commits that make sense individually and that
  all compile are preferentially merged into master.
• PRs with disorganized commits are very likely to be squash-rebased.

Reminder to reviewers

• Read this Checklist
• Put a milestone if possible
• Check labels

[affeldt-aist]

We could maybe rename EVT_max/min to EVT_maxR/minR? (and later the generalization compact_EVT_max to EVT_max) @CohenCyril @yosakaon

[affeldt-aist]

Also, there is this comment left in the code:

(* TODO : Filter not infered *)

What does it mean? Can we remove it? @drouhling