OClCollMeshEngn: perf profile and print kernel exec durations

stimBuffApis/livoxGen1/pcloudStimulusBuffer.cpp

@@ -164,6 +164,12 @@ public:
 			std::cout << __func__ << ": Successfully compacted and collated frame" << std::endl;
 		}
 
+		// Print kernel execution durations
+		auto compactDuration = stimBuff.openClCollatingAndMeshingEngine.getCompactKernelDuration();
+		auto collateDuration = stimBuff.openClCollatingAndMeshingEngine.getCollateKernelDuration();
+		std::cout << __func__ << ": compactKernelDuration=" << compactDuration.count()
+			<< "ms, collateKernelDuration=" << collateDuration.count() << "ms" << std::endl;
+
 		callOriginalCallback();
 	}
 };