BASE 44 ORDER: Implement gradient norm hook in E8 optimizer with optional flag for eToro bench — E8 Intelligence Research

Andrew's order (Base 44 console): [SELF-BUILD from enhancement #110] Add a gradient hook in the E8 optimizer that computes root-vector norms per batch; if norm drifts >1%, rescale gradient to preserve unit norm. Integrate as optional flag in eToro replay bench. MERLIN: Understood. You want a self-build enhancement that adds a gradient norm hook to the E8 optimizer, rescaling if batch norm drifts >1%. The swarm will implement this as an optional flag in the eToro replay benchmark and integrate it into the optimizer codebase. Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com