fix testing with latest Kubernetes

It failed for two reasons:
- CSIDriver v1beta1 as used in csi-driver-host-path v1.3.0 is no longer
  supported.
- GenericEphemeralVolume was removed in Kubernetes master.

prow.sh

@@ -196,7 +196,7 @@ kindest/node:v1.14.10@sha256:f8a66ef82822ab4f7569e91a5bccaf27bceee135c1457c512e5
 # If the deployment script is called with CSI_PROW_TEST_DRIVER=<file name> as
 # environment variable, then it must write a suitable test driver configuration
 # into that file in addition to installing the driver.
-configvar CSI_PROW_DRIVER_VERSION "v1.3.0" "CSI driver version"
+configvar CSI_PROW_DRIVER_VERSION "v1.8.0" "CSI driver version"
 configvar CSI_PROW_DRIVER_REPO https://github.com/kubernetes-csi/csi-driver-host-path "CSI driver repo"
 configvar CSI_PROW_DEPLOYMENT "" "deployment"
 configvar CSI_PROW_DEPLOYMENT_SUFFIX "" "additional suffix in kubernetes-x.yy[suffix].yaml files"
@@ -346,7 +346,7 @@ configvar CSI_PROW_E2E_ALPHA "$(get_versioned_variable CSI_PROW_E2E_ALPHA "${csi
 # kubernetes-csi components must be updated, either by disabling
 # the failing test for "latest" or by updating the test and not running
 # it anymore for older releases.
-configvar CSI_PROW_E2E_ALPHA_GATES_LATEST 'GenericEphemeralVolume=true,CSIStorageCapacity=true' "alpha feature gates for latest Kubernetes"
+configvar CSI_PROW_E2E_ALPHA_GATES_LATEST '' "alpha feature gates for latest Kubernetes"
 configvar CSI_PROW_E2E_ALPHA_GATES "$(get_versioned_variable CSI_PROW_E2E_ALPHA_GATES "${csi_prow_kubernetes_version_suffix}")" "alpha E2E feature gates"
 
 # Which external-snapshotter tag to use for the snapshotter CRD and snapshot-controller deployment
@@ -1304,7 +1304,11 @@ main () {
             delete_cluster_inside_prow_job non-alpha
         fi
 
-        if tests_need_alpha_cluster && [ "${CSI_PROW_E2E_ALPHA_GATES}" ]; then
+        # If the cluster for alpha tests doesn't need any feature gates, then we
+        # could reuse the same cluster as for the other tests. But that would make
+        # the flow in this script harder and wouldn't help in practice because
+        # we have separate Prow jobs for alpha and non-alpha tests.
+        if tests_need_alpha_cluster; then
             # Need to (re)create the cluster.
             start_cluster "${CSI_PROW_E2E_ALPHA_GATES}" || die "starting alpha cluster failed"