Add per-layer MLP type support for executorch export (#18856)

[navsud]

Summary:

Add per-layer MLP type support to the ExecuTorch export path. This allows hybrid models to configure FFN blocks per layer (e.g. skip FFN on specified layers), reducing model size and inference latency.

The per-layer config uses an mlp_type list in ModelArgs, where each layer can be set to "default" (standard FFN) or "skip" (no FFN block). This is extensible to future MLP types.

• Add mlp_type field to ModelArgs (model_args.py) — optional list of strings, one per layer
• Update TransformerBlock.init to accept mlp_type string and skip FFN/ffn_norm creation when mlp_type == "skip" (llama_transformer.py)
• Update TransformerBlock.from_type() to read mlp_type from ModelArgs per layer
• Update TransformerBlock.forward() to pass through attention output directly when mlp_type == "skip"

Reviewed By: ifed-ucsd

Differential Revision: D100682545

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/18856

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

• Rolling out OSDC (ARC) runners on pull workflow for PyTorch trunk commits

❌ 1 New Failure, 2 Unrelated Failures

As of commit 78d8419 with merge base fe71bd4 ():

NEW FAILURE - The following job has failed:

• pull / unittest / macos / macos-job (gh)
  export/tests/test_target_recipes.py::TestTargetRecipes::test_mv3_model

BROKEN TRUNK - The following jobs failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

• pull / unittest / windows / windows-job (gh) (trunk failure)
  ##[error]The operation was canceled.
• pull / unittest-editable / windows / windows-job (gh) (trunk failure)
  ##[error]The operation was canceled.

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[meta-codesync]

@navsud has exported this pull request. If you are a Meta employee, you can view the originating Diff in D100682545.