Update docs to pin K8s at v1.5.1 and remove helm pull workaround

ATT-Comdev do not currently have an image with rbd for K8s 1.5.2, so
until we can move to our own images we should pin the version of K8s at 1.5.1

Also remove workaround for pulling images from gcr.io with docker-engine.
See: moby/moby#30083

This commit should not be merged until this is resolved in Halcyon.
See: att-comdev/halcyon-kubernetes#42

Change-Id: I85c7aa7ba6067ac53013f373dda08b3fac06625d

doc/source/development-environment.rst

@@ -270,7 +270,7 @@ under use.
 .. code-block:: console
 
     export KOLLA_HELM_VERSION=2.1.3
-    export KOLLA_KUBERNETES_VERSION=1.5.2
+    export KOLLA_KUBERNETES_VERSION=1.5.1
 
 .. end
 
@@ -445,28 +445,7 @@ To test that helm is working you can run the following:
     helm init --client-only
     helm repo update
     helm install stable/memcached --name helm-test
-
-.. end
-
-Tiller, which is Helm's server container for starting helm charts, doesn't
-work correctly as of January 21st, 2017 with docker-engine-1.13.0.  The root
-cause and resolution is being tracked in this ssue tracker:
-https://github.com/kubernetes/helm/issues/1838. The Kolla community
-recommends using the following workaround:
-
-.. path .
-.. code-block:: console
-
-kubectl delete --namespace=kube-system deploy tiller-deploy
-helm init --tiller-image docker.io/port/tiller:v2.1.3
-
-.. end
-
-After waiting for the tiller image to be pulled to the kubernetes cluster:
-
-.. path .
-.. code-block:: console
-
+    # check the deployment has succeded
     helm ls
     # and to check via kubectl
     kubectl get all