The issue is that the DeclarativePipeline support on kubernetes-plugin is:

1. Not complete - some options are missing
2. Not aligned with Documentation - by reading docs, I assume I can use all options in declarativePipeline

Example: I'm trying to introduce `slaveConnectTimeout` in my pipeline, because 100 is ridiculously small if you want to pre-warm caches or so. Here is what I get

WorkflowScript: 8: Invalid config option "slaveConnectTimeout" for agent type "kubernetes". Valid config options are [activeDeadlineSeconds, cloud, containerTemplate, containerTemplates, defaultContainer, idleMinutes, inheritFrom, instanceCap, label, nodeSelector, serviceAccount, workingDir, yaml, yamlFile] @ line 8, column 13.
               slaveConnectTimeout 1800

What's actually missing here:

• slaveConnectTimeout Timeout in seconds for an agent to be online.
• name The name of the pod.
• namespace The namespace of the pod.
• nodeUsageMode Either 'NORMAL' or 'EXCLUSIVE', this controls whether Jenkins only schedules jobs with label expressions matching or use the node as much as possible.
• volumes Volumes that are defined for the pod and are mounted by ALL containers.
• envVars Environment variables that are applied to ALL containers.
  • envVar An environment variable whose value is defined inline.
  • secretEnvVar An environment variable whose value is derived from a Kubernetes secret.
• imagePullSecrets List of pull secret names
• annotations Annotations to apply to the pod.

Assuming that `volumes`, `envVars`, `imagePullSecrets` and `annotations` by design have a structure which is hard to read and `name` which looks quite useless here - I'd suggest to add `slaveConnectTimeout`, `namespace` and `nodeUsageMode` to Declarative Pipeline class.

Gonna try to do this by myself, will create the branch/PR