hypothesis-evolution-looplisted

# hypothesis-evolution-loop Goal: - Run the iterative evolution loop from the current persisted run state until convergence. Inputs: - current persisted run artifacts - active `state/STRATEGY_PLAN.json` - optional `resume` flag Outputs: - updated hypotheses - updated island statistics - updated proximity receipt/status artifacts, plus `state/PROXIMITY_GRAPH.json` when the embedding bridge succeeds - updated tournament artifacts - updated convergence counters - updated `state/EVOLUTION_STATE.json` - appended `state/EVOLUTION_ROUNDS.jsonl` - updated `state/PIPELINE_STATE.json` - updated `state/CURRENT_STAGE.json` Context Loading: - Open `skills/shared-references/schema-index.md`. - Read `packages/agent_contracts/pipeline_control.py` before updating `state/EVOLUTION_STATE.json` or `state/COMPLETION_DECISION.json`. - Read `packages/agent_contracts/pipeline_runtime.py` before updating `state/PIPELINE_STATE.json` or `state/CURRENT_STAGE.json`. - Read `packages/agent_contracts/hypothesis.py` before writing any evolved `hypotheses/<id>/HYPOTHESIS.json` artifact. - Read `packages/agent_contracts/state.py` before any island state validation, and read `packages/run_artifacts/island_state.py` before invoking the run-level island persistence helper. - Read `packages/agent_contracts/ranking.py` when the round will write tournament artifacts through downstream ranking work. - Read `packages/agent_contracts/evolution_round.py` before appending `state/EVOLUTION_ROUNDS.jsonl`. - Read t