[KeyboardEvent] Use DOM |code| rather than Windows-based key code

[KeyboardEvent] Use DOM |code| rather than Windows-based key code & introduce XKB support to generate value for DomKey in X11

- Instead of relying on the Windows KeyCode uses Dom |code| to convert to the character. 
- Introduces helper functions to get the meaning of a  DOM |code| in a platform independent way. It supports control characters as well.
It assumes a US keyboard layout is used, so it may only be used when there is no native event or no better way to get the character.
- The table it uses are responsible for converting are: Dom |code| to a printable character, Dom |code| to Dom |key|, Dom |code| to VKeys, and definitions of the xkbcommon keysym type (xkb_keysym_t) and values (XKB_KEY_...) for both xkbcommon and traditional X11.
- To facilitate the Dom |code| to be used, it removes all the code which was used to derive character from the keycode. 

To introduce XKB support in X11.
- XKB related files are moved out of Ozone directory to ui/events/keycodes and now applies to both Ozone and X11. 
- Due to this move, layout_util is removed and xkb_keyboard_code_conversion functionality is moved to ui/events/keycode.
- To derive meaning for X11 events updates returns a dom |key| and character from an XEvent.
- Return the dom |key| determined by the XKB layout result (keysym, character) and character from an XEvent

Originally based on Kevin's CLs.

R=kpschoedel, Wez
BUG=444048

[Habib Virji, 2015-02-25 15:08:51]

It is based on the Kevin's patches. 
https://codereview.chromium.org/939703002/
https://codereview.chromium.org/940553002/
https://codereview.chromium.org/841263005/

[Habib Virji, 2015-03-03 17:24:20]

@kevin/wez: are these changes okay?

[Wez, 2015-03-06 20:17:07]

On 2015/03/03 17:24:20, Habib Virji wrote:
> @kevin/wez: are these changes okay?

Hi Habib, apologies for the delay - can you elaborate on the CL description? 
It's not clear what "non-layout cases" means.

[Wez, 2015-03-06 20:20:56]

On 2015/03/06 20:17:07, Wez wrote:
> On 2015/03/03 17:24:20, Habib Virji wrote:
> > @kevin/wez: are these changes okay?
> 
> Hi Habib, apologies for the delay - can you elaborate on the CL description? 
> It's not clear what "non-layout cases" means.

I'd also like to see a high-level summary of the changes in the CL description -
that in part serves to guide the code-review. e.g. you'd call out the keyboard
code conversion and layout engine API changes, the KeyEvent changes, etc.

[Habib Virji, 2015-03-09 14:25:29]

Thanks for having a look. I have the CL with high level description.

[Habib Virji, 2015-04-20 14:19:55]

@Wez, kevin: Can you please suggest how to proceed on this CL?

[kpschoedel, 2015-04-20 17:07:03]

On 2015/04/20 14:19:55, Habib Virji wrote:
> @Wez, kevin: Can you please suggest how to proceed on this CL?

Part of my CL that this uses breaks some things. I'm working on fixing it this
week.

[Habib Virji, 2015-04-30 13:27:38]

@Kevin: I reckon you have already pushed most of the changes in this patch and
saw you already have other changes in the pipeline. I am closing this CL and
suppose things are already present for me to get DomKey value already there.

[kpschoedel, 2015-04-30 13:59:06]

On 2015/04/30 13:27:38, Habib Virji wrote:
> @Kevin: I reckon you have already pushed most of the changes in this patch and
> saw you already have other changes in the pipeline. I am closing this CL and
> suppose things are already present for me to get DomKey value already there.

With https://codereview.chromium.org/1103263004 KeyEvent::GetDomKey() should
work on all platforms.

[Habib Virji, 2015-04-30 14:01:11]

Is this change correct from that respect:
https://codereview.chromium.org/929053004/patch/40001/50005. Does it need to use
GetDomKey to get the Domkey?

[Habib Virji, 2015-04-30 14:01:13]

Is this change correct from that respect:
https://codereview.chromium.org/929053004/patch/40001/50005. Does it need to use
GetDomKey to get the Domkey?

[kpschoedel, 2015-04-30 14:14:08]

On 2015/04/30 14:01:13, Habib Virji wrote:
> Is this change correct from that respect:
> https://codereview.chromium.org/929053004/patch/40001/50005. Does it need to
use
> GetDomKey to get the Domkey?

That code has been refactored a bit, but that is incorporated in the X11 CL.
After that, ui::KeyEvent::GetDomKey() should be ready for use by blink or other
code.

[Wez, 2015-05-06 01:33:18]

Was this CL supposed to be closed, Habib?

[Habib Virji, 2015-05-06 15:34:46]

On 2015/05/06 01:33:18, Wez wrote:
> Was this CL supposed to be closed, Habib?

Not sure actually, this patch was based on the Kevin patch to support DomKey on
X11. I have seen he has been pushing some of the changes in this patch and so i
have closed my patch. The main aim of this patch was similar to
https://codereview.chromium.org/1103263004, if this lands then DomKey should
work in X11.