Create basic K8s Config

[todaywasawesome]

Ok, we have a very basic config here that moves the env stuff into a configmap.

To deploy to Kubernetes first create a namespace:

kubectl create ns llama

Then apply the manifests under the /deploy/kubernetes directory with

kubectl apply -k deploy/kubernetes/. -n llama

Expose your service however you would normally do that.

[todaywasawesome]

Closes #15

[todaywasawesome]

@mayankchhabra I could add network model storage support here. I think I would want an image that doesn't include any models to go with it. https://github.com/getumbrel/llama-gpt/blob/master/api/Dockerfile

[mayankchhabra]

Sweet! Thanks for working on this @todaywasawesome! Shall we wait for #19 to be merged first? That will remove the models from the images.

[todaywasawesome]

@mayankchhabra your call, if it were me I would probably merge now while there are lots of people interested in trying it. It'll be easy to update once #19 is done.

[todaywasawesome]

@mayankchhabra it's also easy to update which image is being used by anyone consuming our kustomize so they can swap it out whenever they're ready.

[mayankchhabra]

Fair enough! I haven't been able to test it, but let's roll with it since you've it running. Thanks again for your work!