Proposal: Set threaded to False in sr1.

[polybassa]

If the user doesn't specifiy threaded, we should set it to False in sr1 to eliminate the overhead of a thread creation

@gpotter2 What do you think?

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 81.64%. Comparing base (3cbb62e) to head (9e91011).
Report is 3 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #4641   +/-   ##
=======================================
  Coverage   81.64%   81.64%           
=======================================
  Files         359      359           
  Lines       86237    86238    +1     
=======================================
+ Hits        70405    70407    +2     
+ Misses      15832    15831    -1     

Files with missing lines	Coverage Δ
scapy/supersocket.py	73.86% <100.00%> (+0.07%)	⬆️

... and 7 files with indirect coverage changes

[gpotter2]

This was a hard choice but I don't think changing it is good idea. It misses more packets under heavy loads. This code should be entirely rewritten, but that's for another time

[polydroi]

I just had sr1 in mind, since it would only send one packet and immediately stop the send loop. My assumption here is that we would not see the same issue as we had on sr, since after sending we would have the main thread immediately executing the recv loop

[gpotter2]

Hm yeah I see your point. For sr1 I guess it would make sense.