Introduce 8da4w quant for decoder-only text models

[guangy10]

Initial efforts to introduce quantization to native Hugging Face models that are already supported in optimum-executorch. Start with decoder-only text models using "8da4w" for linear layers and int8 for embedding.

Experiment the quantization configs with the following models:

• Qwen3-0.6B
• gemma-3-1b
• HuggingFaceTB/SmolLM2-135M

Example usage

via optimum-cli
optimum-cli export executorch --model Qwen/Qwen3-0.6B --task text-generation --recipe xnnpack --use_custom_sdpa --qlinear --qembedding --output_dir qwen3_8da4w_8we

or use the ExecuTorchModelForCausalLM.from_pretrained.

et_model = ExecuTorchModelForCausalLM.from_pretrained("./qwen3_8da4w_8we")

.pte size comparison:

qwen3_8da4w_8we:
total 1035336
-rw-r--r--  506M  May  1 13:09 model.pte

qwen3_8da4w:
total 1944584
-rw-r--r--  950M  May  1 13:16 model.pte

qwen3_float16:
total 2937408
-rw-r--r--  1.4G  May  1 13:22 model.pte

qwen3_float32:
total 5873128
-rw-r--r--  2.8G  May  1 13:26 model.pte

[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.

[guangy10]

@kimishpatel @metascroy @jerryzh168 for review.

[guangy10]

Tagging @tarun292 for review as we start adding quantization recipe for native HF models.

[guangy10]

Fixed the executorch version check issue on Linux, which returns '0.6.0+cpu', causing parse(executorch_version.__version__) > parse("0.6.0") tests True for executorch==0.6.0. To me it looks like a bug.

[jerryzh168]

quantization related code LGTM