[k8s] refactor gpu labeler script / gpu labeler optionally accepts a context

[SeungjinYang]

1. Use existing helper functions where applicable instead of re-doing a lot of custom logic within the script, which prepares this script to eventually be run on the server as a result of an API call.
2. Allow the script to be called on a specific context, not just the active context.

UX:
if --context is specified

$ python -m sky.utils.kubernetes.gpu_labeler --context syang@my-cluster-2.us-east-1.eksctl.io                      
Found 1 GPU node(s) in the cluster
Using default RuntimeClass for GPU labeling.
Created GPU labeler job for node ip-192-168-56-224.ec2.internal
GPU labeling started - this may take 10 min or more to complete.
To check the status of GPU labeling jobs, run `kubectl get jobs -n kube-system -l job=sky-gpu-labeler --context syang@my-cluster-2.us-east-1.eksctl.io`
You can check if nodes have been labeled by running `kubectl describe nodes --context syang@my-cluster-2.us-east-1.eksctl.io` and looking for labels of the format `skypilot.co/accelerator: <gpu_name>`. 

If --context is not specified

$ python -m sky.utils.kubernetes.gpu_labeler                                                
Found 1 GPU node(s) in the cluster
Using default RuntimeClass for GPU labeling.
Created GPU labeler job for node ip-192-168-56-224.ec2.internal
GPU labeling started - this may take 10 min or more to complete.
To check the status of GPU labeling jobs, run `kubectl get jobs -n kube-system -l job=sky-gpu-labeler`
You can check if nodes have been labeled by running `kubectl describe nodes` and looking for labels of the format `skypilot.co/accelerator: <gpu_name>`. 

Tested (run the relevant ones):

• Code formatting: install pre-commit (auto-check on commit) or bash format.sh
• Manual test: run gpu labeler script on non default context using the --context flag
• Manual test (bw compatibility): run gpu labeler script on default context without using the --context flag
• Manual test: run gpu labeler script using --context flag and make sure the initial cleanup is executed correctly

[Michaelvll]

Awesome @SeungjinYang! LGTM