Investigate switching away from ginkgo/gomega

[mitchnielsen]

Summary

Context: https://github.com/PrefectHQ/prefect-operator/pull/34#discussion_r1722360958

although ginkgo/gomega are included in the scaffolding, it became a bit of a burden over time because it required learning a bespoke testing framework (which is quite complex, as you can tell from their docs). I think this is completely fine for now, but there are complexities here that eventually made us wish we had just used Golang's build in testing package, or at most stretchr/testify for simple assert statements. That would have felt much more Go-native, eliminated weird Ginkgo-specific failures, and made it easier for community members to contribute.

Let's try to get to this sooner than later as it'll only be harder to replace down the road.

Resources

• https://github.com/stretchr/testify
• https://pkg.go.dev/testing

Acceptance criteria

• [ ] Open a PR to demonstrate what it would look like to replace ginkgo/gomega
• [ ] If we like how that goes, open an issue to track replacing the rest of the instances

[mitchnielsen]

Adding Guidry's request:

I'd be down for that, although whichever tool we use should have some kind of watch tool like ginkgo watch or pytest -f