Assorted fixes and optimizations in sage-combinat (mostly partitions and tableaux)

[darij]

• Improve documentation of promotion operators (there are three of them in Sage) on SSYTs.

• Add doc to Lascoux-Schützenberger action on tableaux.

• Add a few more @combinatorial_map decorators on tableaux, partitions, posets.

• Fix doc of the Greene-Kleitman partition method on finite posets.

• Speed up is_column_strict and three further methods on tableaux. PSA: don't use copy.deepcopy unless you really don't know the internal structure of the objects you are copying.

• Fix a bug in the iterator of Partitions(..., outer=a) that caused a to be mutated occasionally, and that required a to be a list (as opposed to, more reasonably, a partition or tuple):

  sage: a = [4,2,1,1,1,1,1]
  sage: for p in Partitions(8, outer=a, min_slope=-1):
      print p
  ....:     
  [3, 2, 1, 1, 1]
  [2, 2, 1, 1, 1, 1]
  [2, 1, 1, 1, 1, 1, 1]
  sage: a
  [3, 2, 1, 1, 1, 1, 1]
  

[git]

Branch pushed to git repo; I updated commit sha1. New commits:

​3f7cec5	Merge branch 'public/combinat/tableaux-et-al-doc' of git://trac.sagemath.org/sage into morestat
​c9d051f	minor improvements to semistandardness testing

[git]

Branch pushed to git repo; I updated commit sha1. New commits:

​dd76d7d	floor and ceiling mutability for integer lists fixed
​045ada0	remove deprecated and hideously confusing list function

[VivianePons]

Hi Darij, thank you very much for this work!

There is something wrong with the documentation of symmetric_group_action_on_values. I must say that I didn't know this operation, so I don't know its definition. But from the code, it seems that your documentation doesn't quite correspond to what is done in the code. When you write this:

        Then, let `a` be the number of opening
        parentheses in the word, and `b` the number of closing
        parentheses (notice that all opening parentheses are left of
        all closing parentheses). Replace the first `b` parentheses
        by the letters `i+1`, and replace the remaining `a` parentheses
        by the letters `i`.

Indeed, the action of s_1 on [[1,2]] should give [[2,1]] which is not a semi-standard tableau... (you obtain the unmatched parenthesizes )(). The code answers [[1,2]], I must say that it is quite obscure and I don't understand what they're doing. But obviously, the documentation does not explain it correctly. Do you have the paper reference where this operation is described?

Apart from that, it seems good to me.

[git]

Branch pushed to git repo; I updated commit sha1. New commits:

​be05e3c

fix the documentation blunder Viviane found

[darij]

Thank you -- this should be fixed now.

[VivianePons]

Well, this is mostly doc improvement. I read everything and it seems quite fine for me. I run the tests on every changed files and built the documentation.

[darij]

Thanks for the review!