fix for ct install getting stuck when printing logs: close readers when done with reading

[skylenet]

What this PR does / why we need it:

In some cases kubectl logs .. was hanging for us. This mainly happened with pods that logged lines with a lot of characters.

The fix of closing both pipes after reading them successfully did the trick and we weren't getting stuck anymore.

---

Which issue this PR fixes *:

There are some closed issue related to this problem:

• Log output for container in 'ct install' does not terminate #332
• Sometimes when printing container logs ct will hang #471

Special notes for your reviewer:

This is how you can reproduce the stuck kubectl logs:

git clone [email protected]:ethpandaops/ethereum-helm-charts.git 
cd ethereum-helm-charts
git checkout bbusa/grandine
docker run --rm --network host --workdir /workdir --volume "$HOME/.kube/config:/root/.kube/config:ro" --volume "$(pwd):/workdir" quay.io/helmpack/chart-testing:v3.8.0 ct install --config ct.yaml --charts charts/ethereum-node --helm-extra-set-args="--values=charts/ethereum-node/ci/networks/sepolia.yaml --values=charts/ethereum-node/ci/clients/execution/nethermind.yaml --values=charts/ethereum-node/ci/clients/consensus/grandine.yaml"

I've also tried using the kubectl request-timeout to see if the command would exit, but that didn't happen. The only way for it to terminate was to close the io readers.

Example of a command that got stuck :

kubectl --request-timeout=1m0s logs ethereum-node-si0kgvb2xz-beacon-0 --namespace ethereum-node-si0kgvb2xz --container grandine

[skylenet]

friendly ping @davidkarlsen @cpanato 🙏

[skylenet]

Any chance on having someone look at this @davidkarlsen ?

[davidkarlsen]

hi @skylenet - sorry for the delay, merging now. Thanks!