This is necessary to catch all cases where locals are referenced before
assignment.  We still keep the _0, _1, _2 versions of LOAD_FAST to help
reduced the byte code size in RAM.

Addresses issue #457.

Small fixes to get it compiling with ARMCC.  I have no idea why
category_t was in the enum definition for qstrs...

This helps with handling "recursive" imports in sane manner, for example
when foo/__init__.py has something like "from foo import submod".

unix: Fix GC not tracing .data

I'm pretty sure these are never reached, since NOT_EQUAL is always
converted into EQUAL in mp_binary_op.  No one should call
type.binary_op directly, they should always go through mp_binary_op
(or mp_obj_is_equal).

With new order of evaluation of defaults, creating the tuple was done in
the wrong spot.

Per https://docs.python.org/3.3/reference/import.html , this is the way to
tell module from package: "Specifically, any module that contains a __path__
attribute is considered a package." And it for sure will be needed to
implement relative imports.

This is alternative implementation of supporting conditionals in qstrdefs.h,
hard to say if it's much cleaner than munging #ifdef's in Python code...

This reverts commit acb133d1.

Conditionals will be suported using C preprocessor.

This simplifies the compiler a little, since now it can do 1 pass over
a function declaration, to determine default arguments.  I would have
done this originally, but CPython 3.3 somehow had the default keyword
args compiled before the default position args (even though they appear
in the other order in the text of the script), and I thought it was
important to have the same order of execution when evaluating default
arguments.  CPython 3.4 has changed the order to the more obvious one,
so we can also change.

Also convert mp_obj_is_integer to an inline function.

Overall this decreased code size (at least on 32-bit x86 machine).

It regressed a bit after implementing float/complex equality.  Now it
should be improved, and support more equality tests.

It has (again) a fast path for ints, and a simplified "slow" path for
everything else.

Also simplify the way str indexing is done (now matches tuple and list).