majentik/Qwen3.5-27B-TurboQuant-MLX-2bit

Remove links to retired duplicate brand variants

Card accuracy pass 2: remove unmeasured speed claims, honest brand labels

Card accuracy sweep: honest brand labeling, remove dead links, upstream KV tip

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

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

Add MLX 2-bit quantized Qwen3.5-27B with TurboQuant KV cache

initial commit