Add CSSTiming to collect aggregate PLT-level stats about CSS.

third_party/WebKit/public/web/WebPerformance.h

Index: third_party/WebKit/public/web/WebPerformance.h
diff --git a/third_party/WebKit/public/web/WebPerformance.h b/third_party/WebKit/public/web/WebPerformance.h
index 7c64df1610be7e13bd2afb218009d43ed6f72f6b..1c148944ef73e1a6a822b429854df3a2e39d3e6d 100644
--- a/third_party/WebKit/public/web/WebPerformance.h
+++ b/third_party/WebKit/public/web/WebPerformance.h
@@ -98,6 +98,9 @@ class WebPerformance {
   BLINK_EXPORT double parseBlockedOnScriptExecutionFromDocumentWriteDuration()
       const;
 
+  // In microseconds (seconds are too coarse).

[Bryan McQuade, 2016/11/04 14:11:30]

are seconds really too coarse if you are using double precision? reading
https://en.wikipedia.org/wiki/Double-precision_floating-point_format it seems
like you have plenty of precision in a double.

[Charlie Harrison, 2016/11/04 16:05:19]

As discussed offline, I think this really has to do with microsecond clamping
done in PerformanceTiming. I'll try to fix it up.

+  BLINK_EXPORT double authorStyleSheetParseDurationBeforeFCP() const;
+
 #if BLINK_IMPLEMENTATION
   WebPerformance(Performance*);
   WebPerformance& operator=(Performance*);