quick fix for SF#1601812 where an error of exactly 0 (very rare) in FLAC__lpc_compute_lp_coefficients() could cause an infinite loop later in FLAC__lpc_quantize_coefficients()

patch: support for borland c++ compile (SF#1599018: http://sourceforge.net/tracker/index.php?func=detail&aid=1599018&group_id=13478&atid=313478)

finished integerized flavor of library; sections with floating point are either #ifdef'd out or written in fixed-point

first pass at making an integer-only flavor of the libraries.  move FLAC__real out of ordinals.h to src/libFLAC/include/private/float.h, add FLAC__double and FLAC__float and use these everywhere instead of double and float, and don't typedef FLAC__real/float/double when building in integer-only mode.  still need to provide integer substitutes in several places.

add wide (64-bit) versions of the FIR filter and inverse filter, remove unused arg from quantizing routine

use double instead of FLAC__real in some places internally; add explicit casts for some implicit cases

fix rare bug in FLAC__lpc_quantize_coefficients where the quantized coeff could get larger than the max allowed value