ENH: add optional sort_order keyword to concat().

[DriesSchaumont]

Is your feature request related to a problem?

xref #33634

ser1 = pd.Series(["1","3"], index=pd.Index(["a", "c"], dtype="object"), dtype="object", name="test1")
ser2 = pd.Series([3, 4, 5], index=pd.Index(["a", "b", "c"], dtype="object"), dtype="float64", name="test2")
ser3 = pd.Series([6,7,8], index=pd.Index(["a", "b", "c"], dtype="object"), dtype="float64", name="test3")
pd.concat([ser1, ser2, ser3], axis=1, sort=False, keys=["test1", "test2", "test3"])
  test1  test2  test3
a     1    3.0    6.0
c     3    5.0    8.0
b   NaN    4.0    7.0

When concatenating, concat uses the index from the first series to determine the order of the non-concatenated axis.
It would be neat to have a way specify this order with either a function or a list. Currently, sort=True can be used, but it only allows for sorting.

Describe the solution you'd like

The solution could be either to:

• add a new argument
• expand the sort argument to take functions or lists
• use the key argument for this.

API breaking implications

I think this change can be backward compatible

Describe alternatives you've considered

As an alternative, reindexing could be used. But this requires the extraction of the largest index first.

[jreback]

we already have .reindex
-1 on additional keywords

[rhshadrach]

-1 as well, I think of combining frames and sorting the index/values are two different operations, it would involve too much complexity combining both into one.

[jreback]

this request is similar to the join_axes argument which was deprecated and removed

this is easily accomplished by a reindex