zfs: default zfs_dmu_offset_next_sync to 0 to avoid data corruption

[emaste]

Summary

This greatly reduces the probability of encountering a data corruption
issue.

From zfs(4), zfs_dmu_offset_next_sync's description:

    Enable forcing TXG sync to find holes.  When enabled forces ZFS
    to sync data when SEEK_HOLE or SEEK_DATA flags are used allowing
    holes in a file to be accurately reported.  When disabled holes
    will not be reported in recently dirtied files.

See https://github.com/openzfs/zfs/issues/15526 for more details.

Sponsored by:   The FreeBSD Foundation

[mav]

It may reduce probability, but it does not fix the problem, plus it makes lseek() to lie, that may or may not be good, depending on situation. I don't think this worth extra activity right now if https://github.com/openzfs/zfs/pull/15571, hopefully fixing the issue, land next days.

[emaste]

I think lseek missing holes isn't that big of a concern given that's the mode we operated in for a very long time, but I agree it looks like 15571 is close to landing and now agree we should just bring that in when ready.