Setup performance: very little concurrency

[Nik-U]

Despite the fact that netmirage-core is multithreaded, setting up a network only maximizes roughly 1.5 cores (with almost all time spent in kernel space). This is because there is a global lock around all network-related netlink calls:
https://elixir.free-electrons.com/linux/v4.15-rc7/source/net/core/rtnetlink.c#L4406

Improving the performance of netmirage-core will ultimately require removing or bypassing this lock. This likely means that we will have to write some code that runs in kernel space and manipulates the routing structures directly. The majority of the netmirage-core logic will need to be moved into this kernel code.