fix: postgres WAL corruption recovery + memory bump + researcher/executor

- Postgres memory: 256M→384M limits, 64M→96M reservations
- Added pg_resetwal -f pre-deploy step to recover from corrupt WAL
  ('PANIC: could not locate a valid checkpoint record' caused by
  force-killed postgres during --force-recreate)
- Added data-checksums initdb arg for future corruption detection
- api→postgres and web→api depends_on: service_healthy→service_started
- Deploy wait loop: fail fast on unhealthy, wait on starting (180s)
- Added researcher/executor to ValidAssignees and frontend dropdowns

.gitea/workflows/deploy.yaml

@@ -211,6 +211,20 @@ jobs:
               set -e
               trap 'rm -f /tmp/nexus-deploy-env' EXIT
               cat > /tmp/nexus-deploy-env
+
+              # ── WAL recovery: reset corrupt WAL that can block postgres startup ──
+              # Force-killed postgres containers can leave stale WAL entries that cause
+              # 'PANIC: could not locate a valid checkpoint record' on next start.
+              # pg_resetwal -f clears the WAL (losing uncommitted tx, which were lost anyway).
+              PG_VOL=\$(docker volume ls -q --filter name=nexus-postgres 2>/dev/null | head -1)
+              if [ -n \"\$PG_VOL\" ]; then
+                echo '🩺 Checking postgres WAL integrity...'
+                docker run --rm -v \"\$PG_VOL:/var/lib/postgresql/data\" \
+                  --entrypoint sh postgres:17-alpine -c '
+                    pg_resetwal -f /var/lib/postgresql/data 2>/dev/null && echo \"✅ WAL reset OK\" || echo \"WAL reset not needed / benign error\"
+                  ' 2>/dev/null || echo 'WAL check skipped'
+              fi
+
               if [ -n '${SERVICE_ARG}' ]; then
                 echo '🚀 Deploying service: ${SERVICE_ARG}'
                 docker compose --env-file /tmp/nexus-deploy-env build ${BUILD_ARGS} ${SERVICE_ARG}