WIP Integrate with munit v1.0.0

[valencik]

This PR is a work in progress to update to munit v1 which should be out "real soon now".

Test are currently failing because of a breaking change in munit v1.

From the munit v1.0.0-M1 release notes:

The FunSuite.{afterEach,afterAll} methods are now evaluated before custom fixtures instead of after custom fixtures. The ordering for beforeAll and beforeEach is unchanged. The diff below demonstrates an example, assuming you have a fixture named myFixture and a test suite named MySuite. In other words, FunSuite.{before*,after*} methods are now evaluated as a regular Fixture[T] that's always the first element of munitFixtures: Seq[Fixture[_]].

The particular failure test is CatsEffectFixturesPlatformSpec as it relies the old behaviour where the fixture's afterAll was called before the suite's afterAll.

With the new behaviour, the fixture's afterAll is the last thing called.
It's not immediately clear to me how we'd adapt this test to the new behaviour.

[armanbilge]

@valencik so I wonder if this is actually a (documentation) bug in munit 🤔 because the scaladoc for afterAll() here specifically says:

/** Runs once after all test cases and after all suite-local fixtures have been tear down. */

[armanbilge]

The particular failure test is CatsEffectFixturesPlatformSpec as it relies the old behaviour where the fixture's afterAll was called before the suite's afterAll.

Ah, I'm so sorry 😂 I obviously didn't read your PR description.

[armanbilge]

It's not immediately clear to me how we'd adapt this test to the new behaviour.

Is this an intended change in MUnit? Why?

[armanbilge]

Once again I failed basic reading test :) this breaking change is deliberate and covered in the release notes.

[valencik]

Closing in favour of #223