How should we test errors in Go? And how does the need for this testing influence the way we construct, handle, and pass around errors within our Golang programs?