Minor performance issue; child restart loop goes one-at-a-time

[rdw]

To repro:

• Fire up Spawning with 50 children: spawning -s 50 wsgi.application

• Wait for them all to launch, with stopwatch in hand.

• Kill its children with pkill -f spawning_child

• Observe that it enters a loop of logs like these:

  (28888) Child died from signal 15 with code 0
  Should start 1 new children
  Controller.spawn_children(number=1)

• Observe how long it takes to do these

It's not a huge performance problem; it only takes 25% longer than the "Controller.spawn_children(number=50)" that happens on launch. Feel free to close this if you feel it's not worth spending time to fix.