So far, don't work for strings as expected.

This implements internal args tuple of arguments, while still keeping
object useful for reporting C-side errors.

Further elaboration is needed.

This is useful because tuple is closest analog of C static array.

vstr is initially intended to deal with arbitrary-length strings. By
providing a bit lower-level API calls, it will be also useful to deal
with arbitrary-length I/O buffers (the difference from strings is that
buffers are filled from "outside", via I/O).

Another issue, especially aggravated by I/O buffer use, is alloc size
vs actual size length. If allocated 1Mb for buffer, but actually
read 1 byte, we don't want to keep rest of 1Mb be locked by this I/O
result, but rather return it to heap ASAP ("shrink" buffer before passing
it to qstr_from_str_take()).

By dispatching to long int methods.

Made sorted() raise an exception instead of aborting when given no arguments; moved around some things in objfun.c as a consequence

This should fix issue #171

Long int is something which doesn't fit into SMALL_INT partion of
machine_int_t. But it's also something which doesn't fit into
machine_int_t in the first place.

We likely should make mp_obj_new_int() inline, and rely on its
encapsulated check rather than inline checks everywhere explicitly.

Also, parser for big small int values is still broken.

Currently it would report "operation not supported" which is confusing. Overall,
this is thinko leading to undefined behavior.