Capture current mission, AI, map and administration development

Include the current addon, terrain integrations, tools, regression tests and recorded development evidence.

docs/evidence/hq-planner-budget-2026-10-02/independent-review.json

@@ -0,0 +1,29 @@
+{
+  "reviewer": "convoy_supply_pool",
+  "status": "approved",
+  "proposal": ".build/gameplay-budget-0219/fix.patch",
+  "focused_test": {
+    "program": "tests/test_hq_planner_budget.py",
+    "source_override": "ERR_HQ_SOURCE=.build/gameplay-budget-0219/addons/core/Scripts/Game/ERR/10_AI/ERR_HQ.c",
+    "result": "PASS",
+    "equivalence_cases": 360,
+    "execution": "Independently executed proposal program with PYTHONPATH=tests, PYTHONDONTWRITEBYTECODE=1, python3 -B and disk-backed TMPDIR=.build/death-diagnostic-2153/test-tmp."
+  },
+  "semantics_review": [
+    "Assigned snapshot is after cleanup, Retask and Investigate; investigation may add an occupied unit before this snapshot.",
+    "Best selection requires no active Assignment or PendingOrder, so each new committed pending order increments total once.",
+    "Inner selection callees construct routes, copy reports, query geometry and buffer journal entries; no synchronous assignment-release or order-delivery callback identified.",
+    "Global occupied units remain distinct from HuntingAssigned; cases exercise active-only, pending-only and both plus independently varied hunting counts."
+  ],
+  "performance_scope": "The observed >10000 versus128 counted unit visits refer only to repeated Assigned scans in the final selection/commit subpass for128 units and64 beliefs with unavailable units. Earlier Assign cleanup, Retask/Investigate and engine time are outside this measurement; no FPS claim.",
+  "mutation": {
+    "change": "Delete the single totalAssigned++ statement; leave snapshot and loop guards intact.",
+    "execution": "Executed once against an isolated temporary ERR_HQ.c copy in disk-backed scratch; canonical target and frozen proposal were unchanged.",
+    "compiled": true,
+    "result": "runtime assertion failure (SIGABRT)",
+    "observed_assertion": "Equal(Setup&, Setup&): a.h.m_Journal.writes==b.h.m_Journal.writes&&a.h.m_Coverage.selections==b.h.m_Coverage.selections&&a.h.m_HomeAvailable.v==b.h.m_HomeAvailable.v",
+    "raw_log_retained": false,
+    "retention_note": "Output was observed in the tool response but not saved as a raw filesystem log. Temporary mutation source and compiled harness were automatically removed. This JSON records the observed result; no rerun was performed to manufacture retained evidence."
+  },
+  "source_changes": "None; only this untracked review record was written. No native launch."
+}