Add cluster creation to CLI

[zipofar]

No description provided.

[gadkins]

Important

• This issue should implement a minimal version of this command.
• The goal is to release this as soon as possible
• GitHub Actions / GitLab CI is the interface we are designing for in this issue
• DO NOT implemented user-friendly error messages (i.e. stack trace dump is fine).
• DO NOT handle edge cases

Base configuration (minimal)

• If no configuration is explicitly passed, create the cluster with the base configuration.
• This base configuration should include an ingress to the virtual cluster's Kubernetes API.
• Set the name and namespace equal to the name the user specifies via --name=[NAME]
• kubeconfig should be saved
• This endpoint should be returned to stdout

For example, a base configuration with on nginx service:

kind: UffizziCluster
apiVersion: uffizzi.com/v1alpha1
metadata:
  name: foo
  namespace: foo
spec:
  ingress:
    class: "nginx"
    host: deleteme2023.uffizzi.cloud
    services:
    - name: default-svc
      namespace: foo
      port: 8080
      ingressAnnotations:
        cert-manager.io/cluster-issuer: "my-uffizzi-letsencrypt"
      certManagerTLSEnabled: true
  ttl: 1h

Command

$ uffizzi cluster create --name=[NAME] --kubeconfig=[KUBECONFIG] --manfiest=[MANIFEST]
  Cluster [NAME] was successfully created.

Conditions

• --name - Required
• --kubeconfig - Required. If it exists, throw error
• --manifest - Optional. Empty cluster if not included

[gadkins]

If --name is not provided, autogenerate a name and return it in the output

I'm moving these details to a separate ticket here.

[axisofentropy]

The behavior if a kubeconfig file is not specified should mirror the behavior of kubectl and other kubernetes tools.

https://kubernetes.io/docs/concepts/configuration/organize-cluster-access-kubeconfig/

By default, kubectl looks for a file named config in the $HOME/.kube directory. You can specify other kubeconfig files by setting the KUBECONFIG environment variable or by setting the --kubeconfig flag.

If a kubeconfig file is found, we should merge the new cluster's connection info into it. This is just YAML, but could be tricky? Best to copy an existing implementation.