Models and data for the paper "Insights from Verification: Training a Verilog Generation LLM with Reinforcement Learning from Testbench Feedback"