The kubernetes-workload scaler should only count running pods (#605)

Signed-off-by: Staffan Olsson <staffan@repos.se>

content/docs/2.6/scalers/kubernetes-workload.md

@@ -3,7 +3,7 @@ title = "Kubernetes Workload"
 layout = "scaler"
 availability = "v2.4+"
 maintainer = "Community"
-description = "Scale applications based on the amount of pods which matches the given selectors."
+description = "Scale applications based on the count of running pods that match the given selectors."
 go_file = "kubernetes_workload_scaler"
 +++
 
@@ -24,6 +24,8 @@ triggers:
 
 > 💡 **Note:** The search scope is limited to the namespace where the `ScaledObject` is deployed.
 
+The count excludes terminated pods, i.e. [pod status](https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.22/#podstatus-v1-core) `phase` equals `Succeeded` or `Failed`.
+
 ### Authentication Parameters
 
 The own KEDA's identity is used to list the pods, so no extra configuration is needed here.