Fix CtsIncidentHostTestCases for RVC AAOS emulator

Same failure in b/161656795, cherry pick the fix

Fix uninitialization issue in JankTracker
We should initialize mSwapDeadline in JankTracker, or it can be a very
large randomized value and then makes jank tracker not collect concrete
jank type of frames appropriately as expected.

Bug: 158029827
Test: run cts -m CtsIncidentHostTestCases -t
   com.android.server.cts.GraphicsStatsValidationTest#testJankyDrawFrame
Change-Id: I057a50a74502918619204f9164f6a954f8e9c5de
(cherry picked from commit 7a0d2247)
(cherry picked from commit 69525945b9f63fe45e2e1cfa7f63320748a3fc25)

libs/hwui/JankTracker.cpp

@@ -139,6 +139,9 @@ void JankTracker::finishFrame(const FrameInfo& frame) {
         (*mGlobalData)->reportJank();
     }
 
+    if (mSwapDeadline < 0) {
+        mSwapDeadline = frame[FrameInfoIndex::IntendedVsync] + mFrameInterval;
+    }
     bool isTripleBuffered = (mSwapDeadline - frame[FrameInfoIndex::IntendedVsync]) > (mFrameInterval * 0.1);
 
     mSwapDeadline = std::max(mSwapDeadline + mFrameInterval,

libs/hwui/JankTracker.h

@@ -75,7 +75,7 @@ private:
 
     std::array<int64_t, NUM_BUCKETS> mThresholds;
     int64_t mFrameInterval;
-    nsecs_t mSwapDeadline;
+    nsecs_t mSwapDeadline = -1;
     // The amount of time we will erase from the total duration to account
     // for SF vsync offsets with HWC2 blocking dequeueBuffers.
     // (Vsync + mDequeueBlockTolerance) is the point at which we expect