As we've already reserved space in the journal this optimization doesn't
actually buy us anything, and when doing list_journal debugging it
deletes information we want.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>