Test that UTF-16LE and UCS-4LE work properly.

Test that UTF-16LE and UCS-4LE work properly.

Those were added to added to newlib in https://codereview.chromium.org/70163008/, and regular UTF-16/UCS-4 were kept as-is since they don't increase pexe size even when used.

R=dschuff@chromium.org
TEST= ./scons run_iconv_test -j32 bitcode=1

Committed: https://src.chromium.org/viewvc/native_client?view=rev&revision=12403

[JF, 2013-11-14 03:32:24]

I'll of course need to roll the newlib version in this CL, but 70163008 needs to
go in first and the toolchain needs to get built.

I'm putting this CL out for review in anticipation of the above, and will
update/trybot before commit.

[Derek Schuff, 2013-11-14 17:13:33]

lgtm

[JF, 2013-11-18 17:23:58]

Waiting on newlib roll:
  https://codereview.chromium.org/75453002

[JF, 2013-11-18 19:31:17]

On 2013/11/18 17:23:58, JF wrote:
> Waiting on newlib roll:
>   https://codereview.chromium.org/75453002

Waiting on this instead:
  https://codereview.chromium.org/75563002

I'll also need to mark the test as only working for PNaCl, other toolchains
can't be rolled at the moment.

[JF, 2013-11-18 20:45:00]

> Waiting on this instead:
>   https://codereview.chromium.org/75563002
> 
> I'll also need to mark the test as only working for PNaCl, other toolchains
> can't be rolled at the moment.

The Newlib change has now rolled into PNaCl, and I marked the test as passing
only for PNaCl. Unrelated breakage prevent me from rolling this for NaCl for
now, but PNaCl is unaffected. I'm now running trybots and will commit this if
they all pass.

[JF, 2013-11-18 22:08:54]

Valgrind failures are the usual ones, I'll commit once the tree opens.

[JF, 2013-11-18 22:36:25]

Committed patchset #3 manually as r12403 (presubmit successful).