Multi channel audio merging

[flashno]

Previous versions of WhisperKit was only using the first channel of Audio, regardless of whether it contained 2 or more channels. This update allows developers to combine all channels available, or choose specific channels to combine to feed into the transcription. To set the configuration on load of WhisperKit, you can pass it through to WhisperKitConfig similar to this:

        let config = WhisperKitConfig(
            ...
            audioInputConfig: AudioInputConfig(channelMode: .sumChannels([1, 3, 5]))
        )

This will sum the 2nd, 4th, and 6th channel (the first channel is index 0).
Another way to pass the channelMode is to directly feed it into the loadAudio function like this:

        let audioArray = try AudioProcessor.loadAudioAsFloatArray(
            fromPath: audioPath,
            channelMode: .specificChannel(0)
        )

This option has been added to all audio processing API's that would read audio.

⚠️ Important: This PR changes the default behavior to now sum and normalize all channels if the audio is not mono to begin with.
The audio merging algorithm works like this:
We find the peak across all channels, check if the peak of the mono (summed) version is higher than any of the peaks of the channels, then we multiply the whole track so that the peak of the mono channel matches the peak of the loudest channel.

Eg: Top mono (merged) buffer, bottom individual channels (pre-merge)

Here you can see how the merged audio maintains the same loudness as the original multi-channel audio file, and you can see the total merged waveform of all the channels.

Resolves these two issues:
#134
#313

[ZachNagengast]

Looks good 👍