docs: upstream-first KV-cache guidance (q8_0/q4_0, mainline Hadamard rotation); fork demoted to experimental

docs: remove AWQ variants (AWQ lane paused; MLX focus)

docs: add status notice

docs: Tier 2 polish — variant matrix + quant trade-off

chore(card): add hardware compatibility section

chore(card): enrich YAML frontmatter (pipeline_tag, language, library_name, inference)

Update KV-cache card with accurate template and fork requirements

Add model card

initial commit