Add testcase for stressing rhlist collisions

[marceloleitner]

As noticed here: https://www.spinics.net/lists/linux-sctp/msg07214.html

This rhlist bug https://www.mail-archive.com/netdev@vger.kernel.org/msg220650.html can impact SCTP in a quite odd way and hard to troubleshoot.

We need a test case stressing this situation.