There are no users of OPT_DATE except for test-parse-options; its
only caller went away in 27ec394a (prune: introduce OPT_EXPIRY_DATE()
and use it, 2013-04-25).

It also has a bug: it does not specify PARSE_OPT_NONEG, but its callback
does not respect the "unset" flag, and will feed NULL to approxidate()
and segfault. Probably this should be marked with NONEG, or the callback
should set the timestamp to some sentinel value (e.g,. "0", or
"(time_t)-1").

But since there are no callers, deleting it means we don't even have to
think about what the right behavior should be.
Signed-off-by: NJeff King <peff@peff.net>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>