chore(typing): added rule 11

[tarekziade]

What does this PR do?

Added Rule 11

forward() must not access non-nn.Module attributes on submodules (breaks pipeline parallelism with Identity replacement).

we want to make sure we just use metadata in config and elesewere when in that function

[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.

[tarekziade]

run-slow: afmoe, cohere2, cwm, dots1, gemma2, gemma3, gemma3n, gpt_oss, minimax, ministral, olmo3, olmo_hybrid, qwen2, qwen2_5_omni, qwen2_5_vl, qwen2_vl

[github-actions]

Workflow Run ⚙️

This comment contains run-slow, running the specified jobs:

models: ["models/afmoe", "models/cohere2", "models/cwm", "models/dots1", "models/gemma2", "models/gemma3", "models/gemma3n", "models/gpt_oss", "models/minimax", "models/ministral", "models/olmo3", "models/olmo_hybrid", "models/qwen2", "models/qwen2_5_omni", "models/qwen2_5_vl", "models/qwen2_vl"]
quantizations: []

[github-actions]

CI Results

Workflow Run ⚙️

Commit Info

Context	Commit	Description
RUN	7b06620a	workflow commit (merge commit)
PR	1da91529	branch commit (from PR)
main	aa1c36f1	base commit (on main)

✅ No failing test specific to this PR 🎉 👏 !

[hmellor]

Can we merge #44699 first?

[github-actions]

[For maintainers] Suggested jobs to run (before merge)

run-slow: qwen3_5, qwen3_5_moe, t5gemma, t5gemma2