[CSS Grid Layout] Add 'auto' fallback for non existent named grid lines

This is a follow up of r174021 where we added support for placing grid items
using named grid lines. We left unfixed the case of the author using non
existent named grid lines. The truth is that it was already partially fixed
by some previous changes.

Instead of forcing resolveGridPositionFromStyle() to return a non resolved
grid position (like 'auto') we try to detect the usage of a non existent
named grid line before trying to resolve the grid position.

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=176022

[svillar, 2014-06-04 15:17:53]

I am not adding any additional test case because we're already covered by
existing tests. Actually we even have a dedicated test case for this fallback
called grid-item-bad-named-area-auto-placement.html

[Julien - ping for review, 2014-06-11 22:24:14]

lgtm

https://codereview.chromium.org/309393002/diff/20001/Source/core/rendering/st...
File Source/core/rendering/style/GridResolvedPosition.cpp (right):

https://codereview.chromium.org/309393002/diff/20001/Source/core/rendering/st...
Source/core/rendering/style/GridResolvedPosition.cpp:25: static inline const
String implicitNamedGridLineForSide(const String& lineName, GridPositionSide
side)
Not sure if returning a const gives us much, especially with C++11 rvalue
coming.

https://codereview.chromium.org/309393002/diff/20001/Source/core/rendering/st...
Source/core/rendering/style/GridResolvedPosition.cpp:30: static bool
isNonExistentNamedLineOrArea(const String& lineName, const RenderStyle& style,
GridPositionSide side)
We should probably just flip the logic as we are doing a double negative
(ASSERT(!isNonExistentNamedLineOrArea())):

ASSERT(isValidNamedLineOrArea())

[svillar, 2014-06-12 10:25:35]

The CQ bit was checked by svillar@igalia.com

[commit-bot: I haz the power, 2014-06-12 10:26:39]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/svillar@igalia.com/309393002/40001

[commit-bot: I haz the power, 2014-06-12 11:30:55]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  win_blink_rel on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/win_blink_rel/builds/11734)

[commit-bot: I haz the power, 2014-06-12 12:26:41]

Change committed as 176022