FPS 0.3

Following the comments over the last few days, I've implemented and
tagged FPS 0.3. Changes include:

    * Clearer docs. Emphasis on support only for latin-1 
    * Partition out the few Word8 functions
    * Merge in all of Simon's QC tests into a combined test/benchmark suite
    * Stress testing on huge data quantities
    * Space profiling, improved some functions
    * Clarify BSD license everything

I think its about ready to become a ByteSequence (or whatever) library:
a useful (actually, critical, imo) interim until we have a full
Unicode/UTF* layer, and a Storable a => Vector a, set up.  At that point,
the Word8/Char issues resolve themselves.

As usual, info is all here:
    http://www.cse.unsw.edu.au/~dons/fps.html

-- Don