CVE-2024-46840

In the Linux kernel, the following vulnerability has been resolved:

btrfs: clean up our handling of refs == 0 in snapshot delete

In reada we BUGON(refs == 0), which could be unkind since we aren't holding a lock on the extent leaf and thus could get a transient incorrect answer. In walkdownproc we also BUGON(refs == 0), which could happen if we have extent tree corruption. Change that to return -EUCLEAN. In dowalkdown() we catch this case and handle it correctly, however we return -EIO, which -EUCLEAN is a more appropriate error code. Finally in walkupproc we have the same BUG_ON(refs == 0), so convert that to proper error handling. Also adjust the error message so we can actually do something with the information.