LU-16866 tests: Use wait_update to check LNet recovery state

LU-16866 tests: Use wait_update to check LNet recovery state

The monitor thread is somtimes woken up on demand and sometimes sleeps
for one second intervals. This makes it hard to precisely predict how
long we need to sleep for ping counts to update and NIs to be
processed out of recovery.
Use wait_update when checking LNet recovery queues and ping counts.
Additional drop rules are added to tests 210 and 211 because it has
been observed that other test instances may issue pings to the node
running 210/211 and cause the ping_count to reset. These additional
drop rules ensure that any incoming messages are dropped.

Test-Parameters: trivial
Test-Parameters: testlist=sanity-lnet env=ONLY=210,211,216
Test-Parameters: testlist=sanity-lnet env=ONLY=211,ONLY_REPEAT=100
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: Ief84388222e46c23952af4ad1d098924e73a8598
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51845
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Frank Sehr <fsehr@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>