FF: allow PipeReader to read from overflowBuffer promptly

[mh105]

Right now if stdout gets crowded by multiple printouts, the _overflowBuffer will catch them during the PipeReader.run() method. However, the for-loop of self._fdpipe.readline is dependent on new lines being added to the pipe. This means leftover bytes from the _overflowBuffer won't print out correctly to the runner stdout panel until the next message is printed, which could be never - so they only get printed at termination of the experiment.

This commit adds the ability to put the leftover bytes from _overflowBuffer into _queue as soon as the _queue is freed up, by coupling it into the isAvailable() method. This behavior is much smoother, without adding much blocking, and is consistent with what the existing comment in PipeReader.run() says:

# Put bytes into buffer if the queue hasn't been emptied quick
# enough. These bytes will be passed along once the queue has
# space.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 49.61%. Comparing base (60088ec) to head (194e456).
Report is 18 commits behind head on release.

Additional details and impacted files

@@           Coverage Diff            @@
##           release    #6833   +/-   ##
========================================
  Coverage    49.61%   49.61%           
========================================
  Files          332      332           
  Lines        61167    61173    +6     
========================================
+ Hits         30348    30354    +6     
  Misses       30819    30819           

Components	Coverage Δ
app	∅ <ø> (∅)
boilerplate	∅ <ø> (∅)
library	∅ <ø> (∅)
vm-safe library	∅ <ø> (∅)