Report expected task queueing time via UMA

components/scheduler/renderer/renderer_scheduler_impl.cc

Index: components/scheduler/renderer/renderer_scheduler_impl.cc
diff --git a/components/scheduler/renderer/renderer_scheduler_impl.cc b/components/scheduler/renderer/renderer_scheduler_impl.cc
index c02624efe41caf4f0b62d006dc0eeb7f3611df19..af6b52ea86a59eb49aa33812ba00c598c7fcfff3 100644
--- a/components/scheduler/renderer/renderer_scheduler_impl.cc
+++ b/components/scheduler/renderer/renderer_scheduler_impl.cc
@@ -22,7 +22,7 @@ namespace scheduler {
 namespace {
 // The run time of loading tasks is strongly bimodal.  The vast majority are
 // very cheap, but there are usually a handful of very expensive tasks (e.g ~1
-// second on a mobile device) so we take a very pesimistic view when estimating
+// second on a mobile device) so we take a very pessimistic view when estimating
 // the cost of loading tasks.
 const int kLoadingTaskEstimationSampleCount = 1000;
 const double kLoadingTaskEstimationPercentile = 99;