Note that compiling on OTP 17.0, or later, will produce beam files
(specifically lager_trunc_io) that are incompatible with R16B03-1, or
earlier, due to erlang:is_map/1 using a new beam opcode (156). However
if compiled on R16B03-1, or earlier, will produce beams files that are
compatible with 17.0, and later, and maps will be formatted where
supported by the active OTP release.

The fix for this was merged in R16 somewhere, so this workaround is no
longer needed, and fixes the EQC test from spotting the divergent
behaviour with io_lib.

See also: http://erlang.org/pipermail/erlang-patches/2012-September/003034.html

Various bugs uncovered during this testing were fixed and added to the
test suite.

There are still differences between the erlang io_lib and the trunc_io in
how ~p/~P are handled, but a first step is useful here.

Bitstrings are binaries that don't end on an 8 bit boundary, so dealing
with the trailing bits is annoying. Thanks to the bit_size bif we don't
have to resort to io_lib for printing them.

As long as the numeral is not the first character, it shouldn't force
quoting on the atom.  For example, abc123 does not need quotes.

Specifically, the handling of lists/binaries/atoms is closer to
io_lib:format's behaviour, and some regular expressions are eliminated.

In string mode, "..", '..' and <<"..">> are stripped, just like with
io:format.

The test if an element in a list was in the printable range didn't
handle the case of a float in the list that compared to be in the
printable range.

This is superior to float_to_list as it only prints the meaningful
digits and 1.0 isn't rendered as "1.00000000000000000000e+00".

This solves an issue reported by 'cloudhead' in IRC where deep lists
weren't being flattened when printed with format specifiers like ~s ~p
and ~P, all of which should flatten the list.