Various Benchmarks that I made

** fasta_main.cpp **
This is the more cleaner version of shorter_main_fixed.cpp. It was my original entry
for "fasta" in benchmarksgame.alioth.debian.org before I made it "shorter". And by shorter,
I mean making single line statements into multiple statement so it fits in 80 character
column limit. It only made my code messy, but it was necessary for it to be accepted.
https://benchmarksgame-team.pages.debian.net/benchmarksgame/performance/fasta.html
https://benchmarksgame-team.pages.debian.net/benchmarksgame/program/fasta-gpp-7.html

** fasta_shorter_main_fixed.cpp **
It is the modified version that is accepted in "fasta" benchmarksgame.alioth.debian.org.
This version was made "shorter" so it can fit on 80 characters per column limit. It
currently is second fastest (Feb 18 2018) while also using way less memory and CPU processing
than the leading implementation (C GCC #2). If it is tested on Windows, it is actually 10%
faster than leading implementation if compiled to 32-bit, and 30% faster when compiled to 64-bit.

** k-nucleotide_main.cpp **
The current leading implementation in k-nucleotide entry. This fixed the fault in
C++ G++ #3 where it was coded differently and naively from top 3 entries that is the cause
of worse performance. This implementation is able to gain 2x performance and marginal
reduction in memory usage. This is explained as a comment in the source code.
https://benchmarksgame-team.pages.debian.net/benchmarksgame/performance/knucleotide.html
https://benchmarksgame-team.pages.debian.net/benchmarksgame/program/knucleotide-gpp-2.html