[InferenceClient] Support billTo header

[Wauplin]

Related to huggingface/huggingface_hub#2940. Discussed in DMs (private link)

Basically adds a parameter allowing a user to be charge all their requests on their organization account instead of personal account.

PR to merge only once "X-HF-Bill-To" header is supported server side cc @SBrandeis

import { InferenceClient } from "@huggingface/inference";

const client = new InferenceClient("hf_token", { billTo: "huggingface" });

const image = await client.textToImage({
	model: "black-forest-labs/FLUX.1-schnell",
	inputs: "A majestic lion in a fantasy forest",
	provider: "fal-ai",
});
/// Use the generated image (it's a Blob)

[Wauplin]

Following server-side update and @SBrandeis 's confirmation that backend is ready, let's merge!