Add support for more input types, add alternative output formats Closes #20

[lbrooney]

Implemented input support for all file formats supported by the user's libsndfile. This means .mp3, flac, and various other formats can now be input and split.

Also added three buttons to the GUI that gives user choice in what output format they want. wav, flac, mp3. Defaults to wav. Using flac and mp3 results in a slightly slower output due to compression, but it saves a lot in file size. I can remove this feature if it's unwanted.

Threw this together quickly today. Based off simple testing this doesn't seem to break any functionality. Just noticed #20 so I guess this closes that. Only changes the GUI, though I'm sure these could be backported to the CLI.

[flutydeer]

Thanks for the contribution, I'll check it out later and consider merging it.

This tool is designed for voice audio files, and mp3 may not be suitable for storing them as it will reduce quality and may introduce some noise. We may need to remind users of this on GUI.

[lbrooney]

Makes sense, should be easy enough to make a conditional label for when mp3 is selected as an output. Trying to think of the best way to word that, something like "MP3 is a lossy format which lowers quality and may introduce unwanted noise."

Should have included this in the original pull request, but here is what the GUI looks like with my additions(minus what we discussed here).