[Ecn-sane] tcp options processing - must should, don't care

[Rodney W. Grimes]

> doesn't look like the ecn related bits were tested.
> 
> https://arxiv.org/pdf/2002.05400.pdf
> 

My guess would be because of "set of minimum TCP requirements that
any TCP speaker MUST implement."  Thus they did NOT look at IP,
nor did they look at SHOULD.  

Now with that said it would make a good paper to do the same
type of investigation for IP and as well extend the coverage
to SHOULD.

> Dave T?ht
-- 
Rod Grimes                                                 rgrimes at freebsd.org