Btrfs: send, don't delay dir move if there's a new parent inode

If between two snapshots we rename an existing directory named X to Y and
make it a child (direct or not) of a new inode named X, we were delaying
the move/rename of the former directory unnecessarily, which would result
in attempting to rename the new directory from its orphan name to name X
prematurely.

Minimal reproducer:

    $ mkfs.btrfs -f /dev/vdd
    $ mount /dev/vdd /mnt
    $ mkdir -p /mnt/merlin/RC/OSD/Source

    $ btrfs subvolume snapshot -r /mnt /mnt/mysnap1

    $ mkdir /mnt/OSD
    $ mv /mnt/merlin/RC/OSD /mnt/OSD/OSD-Plane_788
    $ mv /mnt/OSD /mnt/merlin/RC

    $ btrfs subvolume snapshot -r /mnt /mnt/mysnap2

    $ btrfs send /mnt/mysnap1 -f /tmp/1.snap
    $ btrfs send -p /mnt/mysnap1 /mnt/mysnap2 -f /tmp/2.snap

    $ mkfs.btrfs -f /dev/vdc
    $ mount /dev/vdc /mnt2

    $ btrfs receive /mnt2 -f /tmp/1.snap
    $ btrfs receive /mnt2 -f /tmp/2.snap

The second receive (from an incremental send) failed with the following
error message: "rename o261-7-0 -> merlin/RC/OSD failed".
This is a regression introduced in the 3.16 kernel.

A test case for xfstests follows.

Reported-by: Marc Merlin <marc@merlins.org>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>

fs/btrfs/send.c

@@ -3327,7 +3327,7 @@ static int wait_for_parent_move(struct send_ctx *sctx,
 		if (ret < 0 && ret != -ENOENT) {
 			goto out;
 		} else if (ret == -ENOENT) {
-			ret = 1;
+			ret = 0;
 			break;
 		}