[Bot] Update Inference Providers documentation

[HuggingFaceInfra]

This PR automatically upgrades the @huggingface/tasks and @huggingface/inference packages and regenerates the Inference Providers documentation by running:

cd scripts/inference-providers
pnpm update @huggingface/tasks@latest @huggingface/inference@latest
pnpm run generate

This PR was automatically created (and merged) by the Update Inference Providers Documentation workflow.

---

Note

Low Risk
Documentation-only updates to example model IDs in generated markdown; no runtime or API behavior changes in this repo.

Overview
Regenerated Inference Providers docs after bumping @huggingface/tasks and @huggingface/inference. The diff only updates InferenceSnippet providersMapping example models in provider and task pages—no API spec or template logic changes.

Chat / VLM: Scaleway’s LLM example switches from openai/gpt-oss-120b to meta-llama/Llama-3.1-8B-Instruct; zai-org VLM examples move from GLM-4.6V-Flash to GLM-4.5V (provider pages and chat-completion, image-text-to-text).

Vision tasks: HF Inference image-segmentation examples use jonathandinu/face-parsing instead of mattmdjaga/segformer_b2_clothes (provider + task docs).

Image-to-image: fal-ai points at fal/Qwen-Image-Edit-2511-Multiple-Angles-LoRA (was a face-swap model id); Replicate uses FLUX.2-klein-9B instead of FLUX.2-dev on provider and aggregated task pages.

^{Reviewed by Cursor Bugbot for commit 5025368. Bugbot is set up for automated code reviews on this repo. Configure here.}