Add standard collation UNICODE

The SQL standard defines several standard collations.  Most of them are 
only of legacy interest (IMO), but two are currently relevant: UNICODE 
and UCS_BASIC.  UNICODE sorts by the default Unicode collation algorithm 
specifications and UCS_BASIC sorts by codepoint.

When collation support was added to PostgreSQL, we added UCS_BASIC, 
since that could easily be mapped to the C locale.  But there was no 
straightforward way to provide the UNICODE collation.  (Recall that 
collation support came several releases before ICU support.)

With ICU support, we can provide the UNICODE collation, since it's just 
the root locale.  I suppose one hesitation was that ICU was not a 
standard feature, so this would create variations in the default catalog 
contents, or something like that.  But I think now that we are drifting 
to make ICU more prominent, we can just add that anyway.  I think being 
able to say

     COLLATE UNICODE

instead of

     COLLATE "und-x-icu"

or whatever it is, is pretty useful.

So, attached is a small patch to add this.