[racket] Prince Kuhio Day Failure Followup

In order to avoid problems similar to the failure that caused
DrRacket to fail to start up on 3/26, we've added new tests to
DrRacket. DrDr (http://drdr.racket-lang.org) will now, after
every push, run a script to start up DrRacket multiple times,
once for each different easter egg splash screen. The script sets
up a mode that causes the splash screen to think that today is
one of those special days and then the script checks that
DrRacket started up properly (You can try this yourself by
running collects/tests/drracket/easter-egg.rkt from the git
version of Racket.)

Also, as we prepare for each release, we'll start up DrRacket 366
times, once for each day in the coming year. This we'll do via warp
(http://www.research.att.com/~gsf/warp/warp.html), so we can be
sure that there are no lurking bugs on other days. (Unlike this
test, the DrDr-based test is purely Racket-based.) Thanks to
Brian Mastenbrook for help getting warp+racket going.

Finally, we realize that our users rely on Racket and DrRacket to
work for them. We take this responsibility seriously and are working
to avoid problems like this going forward.

Robby, on behalf of the Racket developers