Multiple precision integers

The multiple-precision arithmetic is implemented in the mp.hpp header and does not need to be compiled separately. The unit test depends the Boost unit testing framework; the other programs have no special dependencies. To install Boost, run brew install boost on macOS or sudo apt-get install libboost-dev on Debian, Ubuntu, or derivatives. To build the unit tests on GNU/Linux or macOS, run g++ mp_test.cpp -o test && ./test --log_level=all (or run the included binary for macOS: ./test --log_level=all). factorial and fibonacci can be built in the same way: g++ fibonacci.cpp -o fibonacci && ./fibonacci or g++ factorial.cpp -o factorial && ./factorial, respectively. Again, binaries are included for macOS. To run expmod, run python expmod.py [a] [b] [n].