Improve test diagnostics

[textbook]

The failing states of the tests weren't particularly useful, so:

• expect(foo.length).toBe(bar) -> expect(foo).toHaveLength(bar), because the latter shows you what was in foo if its length isn't bar - see e.g. jest/prefer-to-have-length
• expect(foo[0].bar).toBe(baz) -> expect(foo).toEqual([{ bar: baz }]) again because then the failing test tells you what foo was, rather than "expected baz received undefined" (or crashing with TypeError: Cannot read properties of undefined)
  • Another alternative, which may be useful in some circumstances, would be expect(foo).toHaveProperty([0, "bar"], baz)

From Growing Object-Oriented Software by Nat Pryce and Steve Freeman, © 2010 Nat Pryce, released under CC BY-SA 4.0.

Also:

• \c videorec; broke the test seeding
• Depending on the implementation, await waitForElementToBeRemoved(deleteButton); can fail because the same button is reused by React - wait for there to be only one button instead
• Use role-based selectors to give a11y feedback
• Use user-event interactions to let the tests focus on behaviour
• An idiomatic delete is 204 No Content, as the resource no longer exists to return

[Dedekind561]

LGTM, thanks @textbook!