Fix typo ENABLE_UTF8MB -> ENABLE_UTF8MB4

[mikeshulman]

See discussion at mikeshulman#1.

[taniwallach]

I think both changes are correct.

The fix is needed as the code is currently inconsistent, but it seems that many systems would not notice the bug. I suspect that systems where the database is not configured to default to utf8mb4 for the connection may have issues at present.

I certainly do not encounter any issues on a Docker setup before the patch, and I could put a 4-byte UTF-8 character into a students name.

Why?

1. My database is configured directly to use utf8mb4 for the connection charset, etc. using docker-config/db/mariadb.cnf so the bug causing a failure to call SET NAMES 'utf8mb4'; is not causing trouble.
2. It seems that Perl's DBD when when mysql_enable_utf8 is used (instead of mysql_enable_utf8mb4) does not really make any effort to prevent 4-byte UTF-8 characters from being passed between Perl and the database. Instead, to documentation instructs users of the limited 3-byte mySQL utf8 charset that such checks are to be added by the people writing code which uses the restricted (somewhat broken) utf8 character set.
   • https://metacpan.org/pod/DBD::mysql
   • https://github.com/perl5-dbi/DBD-mysql/blob/master/lib/DBD/mysql.pm
     • Note the lines (emphasis added)

When the C charset is used then you are responsible for 3-byte UTF-8 sequence checks on input perl scalar strings. Otherwise MySQL server can reject or modify the input statement!