Increase timeout in apps test

I'm not sure why, but it seems it takes sometimes ~one minute to delete
the old job, even though the underlying pod just runs `false` and should
terminate immediately. After bumping the timeout, the test passed 25
times in a row, whereas before it failed on the second attempt.

Fixes b/223331353.

Change-Id: Ice1760a9b7d51f47a3d0e052b62f2aa878c79618
GitOrigin-RevId: 86eb62aee3b09498c85d3cb13fae599c80bed02b

src/go/tests/apps/BUILD.bazel

@@ -10,8 +10,6 @@ go_test(
         "//src/go/cmd/synk",
         "@kubernetes_helm//:helm",
     ],
-    # TODO(b/223331353): fix flake
-    flaky = True,
     rundir = ".",
     tags = [
         "requires-access-token",

src/go/tests/apps/apps_test.go

@@ -346,7 +346,7 @@ func testUpdateChartAssignment_WithFixedJob_BecomesReady(t *testing.T, f *kubete
 
 	if err := backoff.Retry(
 		f.ChartAssignmentHasStatus(&ca, crcapps.ChartAssignmentPhaseReady),
-		backoff.WithMaxRetries(backoff.NewConstantBackOff(time.Second), 60),
+		backoff.WithMaxRetries(backoff.NewConstantBackOff(3*time.Second), 60),
 	); err != nil {
 		t.Fatalf("wait for chart assignment to go from Settled to Ready: %s", err)
 	}