* Ignore extraneous error messages outside of one skipped test

* Fix analysis error

...instead of the engine.

* enable lint prefer_asserts_in_initializer_lists

* enable --assert-initializer

Also, add tests to verify that our coverage is actually being tested!

This adds a way to run only a subset of the tests.
(The new flags do the same thing as 'pub run test'.)

`all.dart` is no longer needed. Furthermore, it causes tests to
be skipped, or to silently fail to run anything.

Fixes #7941

* Bump to test `0.12.20`.

Some test `0.12.20` highlights:
  * introduces `expectLater()`  that returns a `Future` that completes when the matcher has finished running
  * deprecates the `verbose` parameter to `expect()` and the `formatFailure()` (to be removed in `0.13.0`)

Otherwise:
  * to keep up w/ the deprecation of `verbose`, removes `widget_tester` API to pass `verbose` flag (alternatively we could suppress the warning for now)

* Update stack manipulation.

* Fix framecount.

...and fix bugs that the tests uncovered.

WRITE TEST FIND BUG

It was causing the flakiness because it was too short. The default timeout is better.

We now produce a more reasonable error message when we're missing the
flutter_test dependency in a test. Also, remove the flutter_tools stack traces
when the engine dies.

Fixes #6187

This ensures that accidental usages of dart:io's
file API don't creep in over time.

This removes direct file access from within flutter_tools
in favor of using `package:file` via a `FileSystem` that's
accessed via the `ApplicationContext`.

This lays the groundwork for us to be able to easily swap
out the underlying file system when running Flutter tools,
which will be used to provide a record/replay file system,
analogous to what we have for process invocations.

This prevents multiple simultaneous runs of the analyzer from stomping
over each other (e.g. multiple runs of 'update-packages'). Certain
long-lived commands (like analyze, run, logs) are exempted once they've
done enough work to be safe from most stomping action.

This still doesn't make us entirely safe from craziness, e.g. if you're
half way through an 'update-packages' run and you call 'git pull', who
knows what state you'll end up in. But there's only so much one can do.

Fixes https://github.com/flutter/flutter/issues/2762

This makes it possible to substitute 'flutter run' for 'flutter test'
and actually watch a test run on a device.

For any test that depends on flutter_test:

1. Remove any import of 'package:test/test.dart'.

2. Replace `testWidgets('...', (WidgetTester tester) {`
      with `testWidgets('...', (WidgetTester tester) async {`

3. Add an "await" in front of calls to any of the following:
    * tap()
    * tapAt()
    * fling()
    * flingFrom()
    * scroll()
    * scrollAt()
    * pump()
    * pumpWidget()

4. Replace any calls to `tester.flushMicrotasks()` with calls to
   `await tester.idle()`.

There's a guarding API that you can use, if you have particularly
complicated tests, to get better error messages. Search for
TestAsyncUtils.