Add batch files for automatic installation on Windows

[glencoe2004]

This commit adds two new files, Windows-installer.bat and Windows-model_conversion.bat, both of which serve to make using llama.cpp on Windows easier.
Windows-installer.bat installs dependencies, such as Python, and Windows-model_conversion.bat converts the 7B model to FP16, and quantizises to 4 bit.
Additionally, as Windows-model_conversion.bat builds llama.cpp with CMake, it should resolve issue #103.
Note that Windows-model_conversion.bat only currently processes 7B.

Occasionally, Windows-installer.bat will throw an "Unable to complete job - 0x80200002" error. This has no impact on the functionality of the bat file.

While the .bat files still have a few issues (tons of harmless errors if you run Windows-installer.bat after venv creation, Windows-model_conversion.bat only converting and quantizing the 7B model, etc.) being able to install llama.cpp automatically makes the project much easier to use than what we currently have.

[ggerganov]

Overall, it's nice to have easy to use scripts for non-technical users, but I think at this point we will both increase the issue noise and also make it harder to maintain the scripts. Let's first get to a more stable version of the project and we can start adding such conveniences