feat: add a resampling feature

[charles-zablit]

This PR adds a resampling capability in whispering.

Adding a "source_sample_rate" with the sample rate of the source will resample the audio to 16 000 Hz.
It supports supersampling and undersampling.

This addresses #32 and will allow jitsi/jigasi#454 to be merged.

Please let me know if there is anything I can do to improve this PR.

[charles-zablit]

I'm not sure why the CI fails :/

[shirayu]

Thank you for the PR.

However, I think the resampling should be done by the library clients.
Therefore, I am not in favor of adding this feature.
Thank you anyway.

Note: CI fails the code formatting is needed.
I'll add the command (#42).

[charles-zablit]

Hello @shirayu,

I understand that this should be done by the client, as it's a cleaner implementation.

However, this feature would allow us to implement Whispering with Jitsi #41

Furthermore, VOSK server, which also allows live transcription, allows the user to specify the sample rate:
https://github.com/alphacep/vosk-server/blob/ce2c39556c5b44744a7e3a08ad964905c1420046/client-samples/python/asr-test.py#L12

Can you please reconsider merging this?

[shirayu]

Merging this pull request itself is easy.
However, adding various features makes maintenance difficult.
I would like to keep this library as simple as possible.
I am sorry, but please kindly understand this.