feat: Harden research system with hierarchical KV, phase profiling, matched budgets, and counterfactual validation

feat(benchmarks): Add comprehensive baseline comparison system with 14 architectures, 17 metrics, Pareto frontiers, and provenance tags

feat(research): Turn Q-TensorFormer into a verified research system with LFS figure assets

feat(benchmarks): add comprehensive 11-model baseline comparison study, interactive dashboard table, and publication-grade docs

Transform Q-TensorFormer into an Information-Value Adaptive Resource Allocation Architecture