docs(models): add Qwen3.8-Flash-Next coverage (#3691)

Add the model coverage page for the 180B hybrid-attention Engram MoE
(fused TileLang sparse QSA, packed THD sequences, model-owned context
parallelism, reshardable PLE DTensor checkpoints), the 100-step HellaSwag
EP64 fine-tuning recipe, the docs navigation entry, and the README
what's-new bullet and model-table row.

Signed-off-by: HuiyingLi <willwin.lee@gmail.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>