add TextInversion for lpw_stable_diffusion.py

[NormXU]

As StableDiffusionLongPromptWeightingPipeline inherits from StableDiffusionPipeline, we can utilize the functions of TextualInversionLoaderMixin directly to convert negative / prompts if any text inversion tokens are available.

# textual inversion: process multi-vector tokens if necessary
if isinstance(self, TextualInversionLoaderMixin):
    prompt = self.maybe_convert_prompt(prompt, self.tokenizer)
    negative_prompt = self.maybe_convert_prompt(negative_prompt, self.tokenizer)

text_embeddings, uncond_embeddings = get_weighted_text_embeddings(
    pipe=self,
    prompt=prompt,
    uncond_prompt=negative_prompt if do_classifier_free_guidance else None,
    max_embeddings_multiples=max_embeddings_multiples,
)

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint.

[patrickvonplaten]

Cool this looks good to me - wdyt @SkyTNT ?

[SkyTNT]

This is already done in lpw_stable_diffusion.py by #3265

[NormXU]

Since it is already done, I will close this PR.