Refactor some id3scan code

[cjcodeproj]

The id3scan tool should have the code split into different modules, separate from the command line tool.

It should be possible for other front-ends to be developered to call the id3scan functionality with a simple API.

It would also make sense to look into a refactor of the code to break down certain operations into separate OOP classes that can allow for substitutions.

Basic functionality of the tool is:

1. Scan directory for suitable files
2. Read suitable files to build taglib objects
3. Build table of objects into an organized structure
4. Build a list of output files
5. Scan destination directory
6. Write out XML streams as needed.

[cjcodeproj]

Additional Notes

It should be possible to replicate the id3scan functionality from the Python shell, similar to how it can work with the medialibrary module.

A user should be able to pass either a directory path, or a list of files to the core code, and have it build the data structures, which can then be written to a filesystem.