BlinkGCPlugin: Support overloaded trace() call from template.

BlinkGCPlugin: Support overloaded trace() call from template.

With the new setup with InlinedGlobalTracingVisitor, trace() functions are
overloaded and may be called from a traceImpl() function in derived classes.
As traceImpl() is a function template, the trace() overload cannot be resolved
at the time of plugin execution and thus the callee of the trace() call becomes
UnresolvedMemberExpr*.

This patch adds support for detecting Base::trace() call if the
"Base::trace" part is UnresolvedMemberExpr. Specifically, the new code finds
a function looking like trace() from the candidates of UnresolvedMemberExpr,
and if there's one, we treat it as a valid Base::trace() call. The call
needs to give exactly one argument named "visitor" to the function.

A few tests are added to test these cases. A stub for InlinedGlobalTracing-
Visitor is added to tests/heap/stubs.h so it can be used from tests.

BUG=none
R=kouhei@chromium.org
CC=zerny@chromium.org, oilpan-reviews@chromium.org

Committed: https://crrev.com/e5ec4fe6e30327ccff5ed9e2ee57495b24f6878a
Cr-Commit-Position: refs/heads/master@{#317768}

[kouhei (in TOK), 2015-02-24 07:46:33]

lgtm

https://codereview.chromium.org/939303005/diff/20001/tools/clang/blink_gc_plu...
File tools/clang/blink_gc_plugin/BlinkGCPlugin.cpp (right):

https://codereview.chromium.org/939303005/diff/20001/tools/clang/blink_gc_plu...
tools/clang/blink_gc_plugin/BlinkGCPlugin.cpp:516: // Check if the passed
argument is named "visitor".
Nit: Should we document this change in CL desc?

[Yuta Kitamura, 2015-02-24 08:35:59]

Thanks!

https://codereview.chromium.org/939303005/diff/20001/tools/clang/blink_gc_plu...
File tools/clang/blink_gc_plugin/BlinkGCPlugin.cpp (right):

https://codereview.chromium.org/939303005/diff/20001/tools/clang/blink_gc_plu...
tools/clang/blink_gc_plugin/BlinkGCPlugin.cpp:516: // Check if the passed
argument is named "visitor".
On 2015/02/24 07:46:32, kouhei wrote:
> Nit: Should we document this change in CL desc?

Done.

[Yuta Kitamura, 2015-02-24 08:36:06]

The CQ bit was checked by yutak@chromium.org

[commit-bot: I haz the power, 2015-02-24 08:37:10]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/939303005/20001

[commit-bot: I haz the power, 2015-02-24 08:56:00]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-02-24 08:56:37]

Patchset 2 (id:??) landed as
https://crrev.com/e5ec4fe6e30327ccff5ed9e2ee57495b24f6878a
Cr-Commit-Position: refs/heads/master@{#317768}