docs: update empirical benchmark scoreboard with verified v18 production numbers

docs: update empirical benchmarks (99.22% fidelity, 0.78% degradation, 1.75GB disk size)

feat: Final Hybrid FP8 + INT4-g16 (99.2204% fidelity, 0.7796% degradation, 1.75GB)

feat: tie_word_embeddings=True for pure embedding model

feat: remote code support for Hybrid FP8 Attn/GDN + INT4-g16 MLP

feat: Hybrid FP8 + INT4-g16 (99.2204% fidelity, 0.7796% degradation, 2.04GB)

feat: support FP8RowScaledEmbedding and W4A8 Group-32 linear in remote code

feat: FP8 Row-Scaled Embeddings + W4A8-G32 Linear (97.3419% fidelity, 2.6581% degradation, 1.73GB)

docs: sanitize KaTeX syntax ($...$ delims, \geq, and variable formatting)

docs: clarify GGUF modality comparison labels to Not Supported (Text-Only)

docs: fix KaTeX equation blocks and sanitize table angle brackets

docs: publish comprehensive Model Card with 7-format comparisons, architecture deep-dive, MRL, and multimodal inference

docs: update empirical milestone metrics for 4-Pillar checkpoint (96.7267% fidelity)

feat: support SVDLowRankW4A8Embedding and PAS FP8 in remote code

feat: 4-Pillar SVD Vocab + PAS FP8 Quantization (96.7267% fidelity, 3.2733% degradation)

docs: update empirical MRL benchmark metrics and clarify CPU validation

empirical: measured live forward inference (96.5900% fidelity, 3.4100% degradation)

fix: align visual pos_embed shape and ensure seamless loadability

docs: clean unicode multiplication symbols in Model Card

docs: expand Model Card with architecture deep-dive, MRL benchmarks, layer breakdown, and hardware profiling

docs: fix KaTeX and Markdown table rendering syntax for fidelity and KL divergence

Add comprehensive Model Card and GGUF/NVFP4 comparative evaluation matrix

Release WeMM-Embedding-2B-Quantized: Native Sub-1.3GB SafeTensors with <1.0% degradation

initial commit