Update Android SitePerProcessHitTest to new Hit Test Api

I missed a few callsites to WaitForChildFrameSurfaceReady which were in Android tests.
This change updates them to use the new HitTestRegionObserver

TEST=TouchSelectionControllerClientAndroidSiteIsolationTest.BasicSelectionIsolatedIframe
SitePerProcessHitTestBrowserTest.CrossProcessTooltipTestAndroid

Bug: 787945
Change-Id: I0abfa8e4c036552ac416909b15398a53a71ee064
Reviewed-on: https://chromium-review.googlesource.com/1102829Reviewed-by: Ken Buchanan <kenrb@chromium.org>
Commit-Queue: Jonathan Ross <jonross@chromium.org>
Cr-Commit-Position: refs/heads/master@{#568059}

content/browser/site_per_process_browsertest.cc

@@ -10257,7 +10257,7 @@ IN_PROC_BROWSER_TEST_F(TouchSelectionControllerClientAndroidSiteIsolationTest,
   // The child will change with the cross-site navigation. It shouldn't change
   // after this.
   child = root->child_at(0);
-  WaitForChildFrameSurfaceReady(child->current_frame_host());
+  WaitForHitTestDataOrChildSurfaceReady(child->current_frame_host());
 
   RenderWidgetHostViewChildFrame* child_view =
       static_cast<RenderWidgetHostViewChildFrame*>(

content/browser/site_per_process_hit_test_browsertest.cc

@@ -1971,7 +1971,7 @@ IN_PROC_BROWSER_TEST_P(SitePerProcessHitTestBrowserTest,
   // than nullptr. If it did, this test would be unnecessary.
   DCHECK(!rwhv_a->GetCursorManager());
 
-  WaitForChildFrameSurfaceReady(b_node->current_frame_host());
+  WaitForHitTestDataOrChildSurfaceReady(b_node->current_frame_host());
 
   // Make sure the point_in_a_frame value is outside the default 8px margin
   // for the body element.