add support FluxFill inpainting pipeline

[eaidova]

What does this PR do?

Support export and inference for https://huggingface.co/black-forest-labs/FLUX.1-Fill-dev

import torch
from diffusers.utils import load_image

from optimum.intel.openvino import OVDiffusionPipeline

model_id = "black-forest-labs/FLUX.1-Fill-dev"


image = load_image("https://huggingface.co/datasets/diffusers/diffusers-images-docs/resolve/main/cup.png")
mask = load_image("https://huggingface.co/datasets/diffusers/diffusers-images-docs/resolve/main/cup_mask.png")

pipe = OVDiffusionPipeline.from_pretrained(model_id)
image = pipe(
    prompt="a white paper cup",
    image=image,
    mask_image=mask,
    height=512,
    width=512,
    guidance_scale=30,
    num_inference_steps=50,
    max_sequence_length=512,
    generator=torch.Generator("cpu").manual_seed(0),
).images[0]
image.save("flux-fill-dev.png")

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you make sure to update the documentation with your changes?
• Did you write any new necessary tests?

[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.

[echarlaix]

LGTM!

[echarlaix]

thanks for the fix!

[echarlaix]

also @eaidova would you mind adding a test for the CLI as well ? https://github.com/huggingface/optimum-intel/blob/main/tests/openvino/test_exporters_cli.py#L67C5-L67C28

[eaidova]

also @eaidova would you mind adding a test for the CLI as well ? https://github.com/huggingface/optimum-intel/blob/main/tests/openvino/test_exporters_cli.py#L67C5-L67C28

tests provided