[ofw][patch] mcast destroy retry count - parameters

[Sean Hefty]

>Is there a reason user-mode clients should not use the same timeouts and retry
>count?  Just use the same code for kernel- and user-mode clients.  What's good
>enough for the kernel is good enough for user-mode clients.  It will keep the
>code easier to maintain, and will keep consistent functionality.

To add to this, we eventually need a single location for managing multicast
groups.  Otherwise, there's no hope of this ever working correctly with multiple
clients of the same group.

- Sean