✨ Support for Featherless.ai as inference provider

[pohnean]

This PR implements support for Featherless.ai as an inference provider for conversional and text-generation tasks.

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

[hanouticelina]

thanks @pohnean for the PR! it works as expected, except for the text generation task (see my comment below).

[hanouticelina]

@bot /style

[github-actions]

Style fixes have been applied. View the workflow run here.

[pohnean]

Hi @hanouticelina, I've merged your changes in, thanks! It's ready for you to review again.

Btw, I've also added an optional api_key parameter to the _prepare_mapping_info so that models can be run when in staging model.

[hanouticelina]

thanks @pohnean for the iteration. I left a few comments — let's avoid changing anything that's not related to the integration of Featherless AI, there is no need to pass the api_key to HfApi in _fetch_inference_provider_mapping, you don't need to be logged in to fetch model info.

[pohnean]

Hi @hanouticelina, issues resolved.

[hanouticelina]

looks good, thanks @pohnean for the PR!

[Wauplin]

Looks great! Thanks both for integration and reviews :)