update_engine: Report VPD write failure to UMA

Currently we see many first actives to come from non-FSI images. But we have not
been able to figure out why. This CL, reports a new error
kFirstActiveOmahaPingSentPersistenceError when writing the first active omaha
flag into VPD fails. This allows us to see if that is the actual cause of the
problem.

CL:1062659 adds the enum value on the Chrome side.

BUG=chromium:833980
TEST=unittests
TEST=precq

Change-Id: I65e233c5f895489ba905494fb20d7b00d0c4af10
Reviewed-on: https://chromium-review.googlesource.com/1062662
Commit-Ready: ChromeOS CL Exonerator Bot <chromiumos-cl-exonerator@appspot.gserviceaccount.com>
Tested-by: Amin Hassani <ahassani@chromium.org>
Reviewed-by: Amin Hassani <ahassani@chromium.org>
Reviewed-by: Ben Chan <benchan@chromium.org>
diff --git a/hardware_android.cc b/hardware_android.cc
index 7958fbf..0e5abaa 100644
--- a/hardware_android.cc
+++ b/hardware_android.cc
@@ -214,9 +214,10 @@
   return false;
 }
 
-void HardwareAndroid::SetFirstActiveOmahaPingSent() {
-  LOG(WARNING) << "STUB: Assuming first active omaha is never set.";
-  return;
+bool HardwareAndroid::SetFirstActiveOmahaPingSent() {
+  LOG(WARNING) << "STUB: Assuming first active omaha is set.";
+  // We will set it true, so its failure doesn't cause escalation.
+  return true;
 }
 
 }  // namespace chromeos_update_engine