We removed all codepaths where s_umount is taken beneath open_mutex and
bd_holder_lock so don't make things more complicated than they need to
be and hold s_umount over block device opening.

CC: Ryusuke Konishi <konishi.ryusuke@gmail.com>
CC: <linux-nilfs@vger.kernel.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20231101172739.8676-1-jack@suse.czAcked-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>