queue-runner: limit parallelism of CPU intensive operations

My current theory is that running more parallel xz than available CPU
cores is reducing our overall throughput by requiring more scheduling
overhead and more cache thrashing.

src/root/common.tt

@@ -270,6 +270,8 @@ BLOCK renderBusyStatus;
     <strong>Sending inputs</strong>
   [% ELSIF step.busy == 30 %]
     <strong>Building</strong>
+  [% ELSIF step.busy == 35 %]
+    <strong>Waiting to receive outputs</strong>
   [% ELSIF step.busy == 40 %]
     <strong>Receiving outputs</strong>
   [% ELSIF step.busy == 50 %]