Revert "tools: add install_yq"

[SamYuan1990]

Reverts sustainable-computing-io/kepler#1393

There seems no need for a PR to add yq. In CI if we have grafana features, which yq installed at https://github.com/sustainable-computing-io/local-dev-cluster/blob/main/main.sh#L190 this covers case which we build up instance on a new created ec2.

and if we need to handle json, yq and jq been installed via GHA naivety. https://github.com/actions/runner-images/blob/7bb1d84f7071bfa9c350d7552d9631d9a69bfdb0/images/ubuntu/Ubuntu2404-Readme.md?plain=1#L72-L86

So, what's the reason we add yq install in our code?

[maryamtahhan]

So, what's the reason we add yq install in our code?

To improve developer experience. If YQ isn't installed then local-dev-cluster will skip the grafana dashboard automation step and the developer will need to manually import the dashboard.

TBH as a developer it would be great if the first thing I could do after cloning the kepler repo is run make deps or some other target that installs all the tools that I need to deploy the full stack. I don't see a place where the local-dev-cluster pre-requisites are automated for developer experience and we rely on the (new) developer to run a bunch of steps to get up and running with kepler.

IMHO - the kepler developer onboarding experience could be greatly improved.

[rootfs]

I tested jq and yq on my end, for the use case kepler needs (i.e. parsing kubectl output), both can do the job. Since we only need one, what about replacing existing jq reference with yq?