[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
Auto-generated documentation and dependency patch only; no runtime or API implementation changes in this repo.

Overview
Regenerates Inference Providers docs after bumping @huggingface/tasks from 0.21.10 to 0.21.11 (lockfile updated; @huggingface/inference unchanged).

Cohere example models in InferenceSnippet blocks are updated: LLM docs now use CohereLabs/tiny-aya-global instead of tiny-aya-earth; VLM docs use CohereLabs/aya-vision-32b with provider id c4ai-aya-vision-32b instead of command-a-vision-07-2025. The same Cohere VLM mapping is applied on the chat-completion and image-text-to-text task pages.

SambaNova gains a Chat Completion (LLM) section on its provider page (openai/gpt-oss-120b) and is added to the multi-provider LLM snippet on the chat-completion task page. In that combined LLM mapping, publicai is no longer included (per the regenerated mapping from the updated tasks package).

^{Reviewed by Cursor Bugbot for commit 2cda263. Bugbot is set up for automated code reviews on this repo. Configure here.}

[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.