remove prune setting for eks cluster

[steved]

AWS-CDK's lambda for generic kubernetes manifests doesn't support 1.22 officially so prune bails on the missing API versions (like ingress). This occurs during updates to an existing cluster. If/when we update Calico, this may affect how the upgrades from one version to another works.

See aws/aws-cdk#19843 for more context. aws/aws-cdk#15736 (comment) does indicate a workaround, but it doesn't seem very palatable.

deployer_image: quay.io/domino/deployer:steved-destroy-deploy.latest

[steved]

Tested with a successful deploy of 1.22 on master. Adding a new unmanaged node group and cdk deploy results in:

Received response status [FAILED] from custom resource. Message returned: Error: b'configmap/aws-auth configured\nerror: error retrieving RESTMappings to prune: invalid resource extensions/v1beta1, Kind=Ingress, Namespaced=true: no matches for kind "Ingress" in version "extensions/v1beta1"\n' Logs: /aws/lambda/steved-cdk-1-EksStackNestedStackEk-Handler886CB40B-2ak0Zp34wmv1 at invokeUserFunction (/var/task/framework.js:96:19) at processTicksAndRejections (internal/process/task_queues.js:97:5) at async onEvent (/var/task/framework.js:19:27) at async Runtime.handler (/var/task/cfn-response.js:48:13) (RequestId: 89089f1c-1801-4430-9479-200c70529c8e)

With this change, it completes successfully.