Data-Driven Approaches for Fatigue Prediction of Additively Manufactured Ti-6AL-4V and Simulation of Surface Defect Effect on Pipelines