Fix panic when kubeconfig is not valid

[TyBrown]

If the kubeconfig is invalid for some reason (misconfigured, typo, etc.), currently kbenv will panic with a stacktrace that doesn't really say what's gone wrong. At the very least, we can handle the underlying error and exit to avoid the stacktrace.

Before:

$ kubectl get pod
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x1e0 pc=0x1c89b0f]

goroutine 1 [running]:
github.com/little-angry-clouds/kubernetes-binaries-managers/internal/helpers.KubeGetVersion(0xc0000b762c, 0x4, 0x1ef5168, 0x1)
	/private/tmp/kbenv-20210729-90323-1ezpqxa/kubernetes-binaries-managers-0.2.6/internal/helpers/kube_client.go:57 +0x10f
github.com/little-angry-clouds/kubernetes-binaries-managers/internal/wrapper.Wrapper(0x1e64889, 0x7)
	/private/tmp/kbenv-20210729-90323-1ezpqxa/kubernetes-binaries-managers-0.2.6/internal/wrapper/main.go:61 +0x77f
main.main()
	/private/tmp/kbenv-20210729-90323-1ezpqxa/kubernetes-binaries-managers-0.2.6/cmd/kubectl-wrapper/main.go:10 +0x36

After:

$ ./kubectl-wrapper get pod
kbenv failed to load kubeconfig: invalid configuration: no configuration has been provided

[alexppg]

Thanks for the contribution and the error report! I've created an issue, which you can see above, in case you'd like to add a test for this. I'll merge it right away anyway.

[alexppg]

There's a release with this changes. Thanks again!

[TyBrown]

Thanks for letting me know about the release! I'll be sure to update when I get back to work on Monday! :) I appreciate your quick response to merging this in and getting it out! :)