foo sound synthesis

changelog

added some examples to Examples/script
started to add examples from KernelReference to Examples/kernel
removed banner printed on startup
added (about) function to print banner
added (version) function to get a list of (major minor subminor) version numbers
nearly rewrote SoundFile.m again (oh well)
discovered some bugs in libsndfile
changed to libsndfile 1.0.10 as minimum requirement
fixed prompt displaying issue in non-readline mode
BandlimittedNoise: added random seed to archiving methods
disabled SoundFile flushing from within Context (SoundFile's responsability)
caught probably long-term bug in task.m related to parameter in (make-task)

fixed bug in ./bootstrap (INSTALL removal)
include GSConfig.h for support of GNUStep fakemain hack on some systems
added --no-rl option and toplevel-no-rl.foo for non-readline operation
read user defaults now from (in this order):
- /etc/foo/init.foo
- - ~/.foo/init.foo
- - ./.init.foo
(retained backwards compatability for ~/.initfoo and ./.initfoo)
added user-accessible (foo-default-play-command)

adopted to fooelk-0.0.2
switch on case-sensitive mode of elk (off by default now in elk)
added custom reader for fractional tones (replaces old semitone reader from customized elk)
readline reader extension/readrationals now in fooelk-0.0.2
restructured source tree
reviewed gnustep-make support
introduced autoconf to guess/set location of fooelk/elk and foo
replaced foo.sh startup script by foo binary

this is the first port of foo to linux. it is mostly compatible to foo-2.1 for NeXTStep®.

last version for NeXTStep®

porting and further development is based on this version of foo