Document BF16 and Q8_0 release tiers

Add Q8_0 generation model

Add BF16 generation model

weights: add yue2-3.6B-v1.0-Q4_K_M.gguf

weights: add sheetsage2-mert2-0.7B-v1.0-F16.gguf

weights: add yue2-semantic-tokenizer-0.7B-v1.0-F16.gguf

weights: add yue2-vae-v1.0-F16.gguf

weights: add yue2-instrumental-cot-full-v1.0-F16-LoRA.gguf

weights: add yue2-realaudio-nar-v9-v1.0-F16-LoRA.gguf

weights: add yue2-qwen.tiktoken

docs: add release checksums

docs: add yuey.cpp model card

initial commit