Improve the handling of inference for instance fields in the task model

Improve the handling of inference for instance fields in the task model
strong mode inference.  This CL adds instance fields to the set of variables
for which dependencies are computed, and re-resolves each instance field
initializer using type information inferred for the static variables upon
which it depends.

This CL also stops static variables with existing type information from having
their types overwritten.

This CL ports over a number of the DDC inference tests, more remain to be ported.

There is still at least one outstanding issue with instance variable inference,
see #354 (and the two tests marked fail added in this CL).

BUG=
R=brianwilkerson@google.com

Committed: https://github.com/dart-lang/sdk/commit/9be2107f391dec2699da031159b819257cd68ab5

[Leaf, 2015-09-25 20:19:51]

leafp@google.com changed reviewers:
+ brianwilkerson@google.com, paulberry@google.com

[Leaf, 2015-09-25 20:21:04]

I'm think this may not be the right long term fix, but it addresses a number of
issues that John and Vijay have been running into.  Take a look, see what you
think!

[Brian Wilkerson, 2015-09-25 20:50:17]

LGTM

With this change are we now duplicating work by re-resolving instance field
initializers somewhere else?

Also, the class InstanceMemberInferrer checks for to see whether a field has an
implicit type, but doesn't take into account the possibility of an inferred
type. Is there any interaction with these changes?

https://codereview.chromium.org/1370793002/diff/1/pkg/analyzer/lib/src/genera...
File pkg/analyzer/lib/src/generated/resolver.dart (right):

https://codereview.chromium.org/1370793002/diff/1/pkg/analyzer/lib/src/genera...
pkg/analyzer/lib/src/generated/resolver.dart:9675:
_addStaticVariables(node.fields.variables);
nit: we should change the name _addStaticVariables given that they aren't
necessarily static.

https://codereview.chromium.org/1370793002/diff/1/pkg/analyzer/test/src/task/...
File pkg/analyzer/test/src/task/dart_test.dart (right):

https://codereview.chromium.org/1370793002/diff/1/pkg/analyzer/test/src/task/...
pkg/analyzer/test/src/task/dart_test.dart:1911: enableStrongMode();
It looks like all of the tests in this class call enableStrongMode(). Consider
moving this to the setUp() method.

[Leaf, 2015-09-26 00:09:25]

On 2015/09/25 20:50:17, Brian Wilkerson wrote:
> LGTM
> 
> With this change are we now duplicating work by re-resolving instance field
> initializers somewhere else?

As best I can tell, instance field initializers were never being re-resolved.

> Also, the class InstanceMemberInferrer checks for to see whether a field has
an
> implicit type, but doesn't take into account the possibility of an inferred
> type. Is there any interaction with these changes?

I don't think we currently do any inference on instance fields before this point
(just a re-resolution), 
so I don't think so.

[Leaf, 2015-09-26 00:09:33]

https://codereview.chromium.org/1370793002/diff/1/pkg/analyzer/lib/src/genera...
File pkg/analyzer/lib/src/generated/resolver.dart (right):

https://codereview.chromium.org/1370793002/diff/1/pkg/analyzer/lib/src/genera...
pkg/analyzer/lib/src/generated/resolver.dart:9675:
_addStaticVariables(node.fields.variables);
On 2015/09/25 20:50:17, Brian Wilkerson wrote:
> nit: we should change the name _addStaticVariables given that they aren't
> necessarily static.

Done.

https://codereview.chromium.org/1370793002/diff/1/pkg/analyzer/test/src/task/...
File pkg/analyzer/test/src/task/dart_test.dart (right):

https://codereview.chromium.org/1370793002/diff/1/pkg/analyzer/test/src/task/...
pkg/analyzer/test/src/task/dart_test.dart:1911: enableStrongMode();
On 2015/09/25 20:50:17, Brian Wilkerson wrote:
> It looks like all of the tests in this class call enableStrongMode(). Consider
> moving this to the setUp() method.

All but the first.  I split all the strong mode ones into a
StrongModeInferenceTest class, since that's what they are anyway.

[Brian Wilkerson, 2015-09-26 00:17:47]

LGTM

[Leaf, 2015-09-28 17:57:13]

Committed patchset #2 (id:10001) manually as
9be2107f391dec2699da031159b819257cd68ab5 (presubmit successful).