fix: increase max_tokens to 2048 and add robust markdown JSON cleaner for NVIDIA SLM

fix: update paymentMethod preset choice from wire_transfer to ach for Gradio component validation

feat: add quick test samples preset bar to TransactionForm dashboard component

fix: execute explanation generation on CPU without 90s GPU scheduling delay

fix: prioritize text json dataset for demo scenarios in app.py

perf: remove top-level GPU invocation to prevent ZeroGPU startup queueing delay

fix: decorate _generate_explanation with @_gpu_decorator for ZeroGPU runtime

fix: register ZeroGPU compatibility startup hook in app.py

fix: default SLM_MODEL_NAME to openai/gpt-oss-20b for NVIDIA Build API

refactor: eliminate Qwen2.5-0.5B-Instruct local GPU loading, use pure NVIDIA Build API or 0-GPU fallback

feat: integrate NVIDIA Build API (openai/gpt-oss-20b) into SLM explanation layer with reasoning support

fix: replace binary dataset with text demo_scenarios.json and enable dynamic RAZOR Q&A

fix: format RAZOR assistant messages with summary, recommended action, and key signals

fix: resolve JSX closing tag syntax error in scenarios page for Vercel deployment

fix: improve server and client error extraction to eliminate HTTP 200 generic message

fix: update Pydantic ExplanationInput defaults, rename assistant to RAZOR, and add null-safety to scenario distribution

fix: resolve TypeError in number formatting and add interactive SLM Co-Pilot chat drawer

feat: implement Vercel Next.js server-side API proxy architecture to eliminate browser CORS

feat: update app.py with user launch configuration and CORS origin logging

fix: restore standard demo.launch loop with strict_cors=False for HF Space persistence

fix: enable FastAPI CORSMiddleware in app.py to resolve cross-origin preflight requests

fix: resolve ZeroGPU CUDA init boundary in model_loader.py

fix: refine SLM vs fallback explanation source display

feat: upgrade UX to live operational merchant risk console with real request tracing

feat: complete Phase 10 Vercel deployment, QA & production demo polish

feat: complete RazorShield Next.js landing page & risk console frontend

fix: shorten README short_description for HF YAML metadata limit

build: track model joblib artifacts with Git LFS for HF Spaces

fix: update merchant state property names in app.py

feat: complete RazorShield AI risk platform