feat: add KRM benchmark example

[bharathkkb]

This adds a KRM benchmark example

• Also moves pkg init and install-resource-group from DefaultInit to DefaultApply for krmt.
  Reasoning for this is because DefaultApply maybe run multiple times. In particular pkg init is not idempotent and will error out if a pkg is already inited.

[morgante]

Before approving this, I'd like to understand a bit more about what we're trying to benchmark. Do we want to benchmark kpt performance, or just kubectl apply behavior?

[bharathkkb]

@morgante We are trying to benchmark KCC actuation time from object creation to resource actuation. In general across tools, we are trying to benchmark time from when a tool is invoked create N resources to when those resources are actually created.

[morgante]

@morgante We are trying to benchmark KCC actuation time from object creation to resource actuation. In general across tools, we are trying to benchmark time from when a tool is invoked create N resources to when those resources are actually created.

Right, what I'm confused on is are we trying to measure kpt or KCC here? We can chat offline.