Update model card: matched 444-decision benchmark vs TypeSafe Jev

docs: pin matched live Jev API benchmark (+13.52% lead on SWE-bench)

docs: clarify self-eval methodology, release checksums, and domain specializations

Upload README.md with huggingface_hub

Upload rl_agent_config.json with huggingface_hub

Upload 3 files

Upload assets/cortex_banner.jpg with huggingface_hub

Upload assets/cortex_logo.jpg with huggingface_hub

Upload calibration_temperatures.json with huggingface_hub

Upload config.json with huggingface_hub

Upload README.md with huggingface_hub

initial commit