Add vram_gb to all rows (artifact weight-file size) — unblocks /lab quality-vs-VRAM chart (protoLab#14, protoContent#394)

Add gates.json — canonical release-gate thresholds for /lab model-compare (protoContent#399)

Ornith-1.0-35B-NVFP4 release: gate + coherence + speed-v2 rows

Agents-A1 release rows: paired gates (both sides), GGUF gate, speed matrices

first rows: Ornith-1.0-9B-NVFP4 release (gate, speed-v2, depth, GGUF matrix, coherence)

initial commit