Docs:Users should be encouraged to use UTF8 for the DB

[jamoore]

In ​PR 171 Andrew fixed the unix text install instructions for PostgreSQL to use UTF8 rather than ASCII. This was brought up on the forums - ​https://www.openmicroscopy.org/community/viewtopic.php?f=5&t=3598

We need to make clear across the board (and on the FAQ?) that UTF-8 is preferred. The Windows instructions need to be similarly reviewed and possibly screenshots re-taken.

Further, Roger brought up the issue of locales and pointed at the docs on pg.org: ​http://www.postgresql.org/docs/9.2/static/locale.html

See ​https://chat.glencoesoftware.com/chatlogs/devteam/2012-12-04.html for the conversation:

[09:56:49] <rleigh> josh: They don't need to drop unless they really want UTF-8. Documenting this would be useful.  Additionally, when you do create with UTF-8, you also get the collation rules for that locale, so there may be differences depending upon the OS and locale in use here.
[09:57:29] <rleigh> E.g. on Debian, this defaults to the system locale (for my systems, that's en_GB.UTF-8, so I get the en_GB collation rules).

[sbesson]

Should be mandatory since OMERO 5.1.0.