Don't allow "" value for kubelet-extra-config

Hello,

When set "" value to kubelet-extra-config it is accepted by the charm, but then agent fails to restart kubelet, which leads to cluster pods going down in 5 minutes.

Environment: kubernetes-worker charm revision 44 from channel 1.24/stable deployed on Bionic machine.

Steps to reproduce:
1. Deploy juju k8s bundle
2. juju config kubernetes-worker kubelet-extra-config=""
3. seeing this in kubernetes-worker logs
```
unit-kubernetes-worker-15: 13:53:21 ERROR unit.kubernetes-worker/15.juju-log Hook error:
Traceback (most recent call last):
  File "/var/lib/juju/agents/unit-kubernetes-worker-15/.venv/lib/python3.6/site-packages/charms/reactive/__init__.py", line 74, in main
    bus.dispatch(restricted=restricted_mode)
  File "/var/lib/juju/agents/unit-kubernetes-worker-15/.venv/lib/python3.6/site-packages/charms/reactive/bus.py", line 390, in dispatch
    _invoke(other_handlers)
  File "/var/lib/juju/agents/unit-kubernetes-worker-15/.venv/lib/python3.6/site-packages/charms/reactive/bus.py", line 359, in _invoke
    handler.invoke()
  File "/var/lib/juju/agents/unit-kubernetes-worker-15/.venv/lib/python3.6/site-packages/charms/reactive/bus.py", line 181, in invoke
    self._action(*args)
  File "/var/lib/juju/agents/unit-kubernetes-worker-15/charm/reactive/kubernetes_worker.py", line 668, in start_worker
    configure_kubelet(dns_domain, dns_ip, registry, has_xcp=kube_control.has_xcp)
  File "lib/charms/layer/kubernetes_common.py", line 1094, in configure_kubelet
    merge_kubelet_extra_config(kubelet_config, kubelet_extra_config)
  File "lib/charms/layer/kubernetes_common.py", line 965, in merge_kubelet_extra_config
    for k, extra_config_value in extra_config.items():
AttributeError: 'NoneType' object has no attribute 'items'
```

The expected behavior would be:
1. Deploy juju k8s bundle
2. juju config kubernetes-worker kubelet-extra-config=""
3. Get an error from `juju config` saying that "" is not allowed value.

The issue's impact was pretty bad in production as we used `juju config kubernetes-worker kubelet-extra-config=""` to reset the experimental kubelet configuration and thought that reset didn't work. So needed to fix kubelet configuration manually, while part of the cluster went down in 5 minutes timeout. Of course we will use only `juju config --reset` from now on. But still would be great to fix configuration validation.
Many thanks in advance!

Best Regards,
Alex.