CLAUDE: add run_infer_local.sh — local GPU infer_server (no DGX/docker/transfer)

Run the PyTorch L1+L2 server on the workstation 5060 Ti; Java pipeline points at 127.0.0.1:5577.
Verified: server loads L1(weighted9_pm_s)+L2(l2_v1) on CUDA, warm-up + pyramid build OK.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

run_infer_local.sh

+#!/usr/bin/env bash
+# run_infer_local.sh - part of imagej_elphel_dnn — Elphel PyTorch DNN companions to imagej-elphel
+#
+# Copyright (C) 2026 Elphel, Inc.
+#
+# -----------------------------------------------------------------------------
+#  imagej_elphel_dnn is free software: you can redistribute it and/or modify it under the terms of the
+#  GNU General Public License as published by the Free Software Foundation, either version 3 of the License,
+#  or (at your option) any later version. Distributed WITHOUT ANY WARRANTY. See <http://www.gnu.org/licenses/>.
+# -----------------------------------------------------------------------------
+# Run the PyTorch infer_server LOCALLY on the workstation GPU (no DGX, no docker, no network transfer).
+# Point the Java pipeline at host:port 127.0.0.1:5577 (DNN remote host). By Claude on 06/27/2026.
+set -e
+VENV="${VENV:-/home/elphel/git/dnn_infer_venv}"   # torch==2.7.1+cu128 GPU venv
+RUN="${RUN:-runs/weighted9_pm_s}"                  # L1 run dir (model.pt)
+RUN2="${RUN2:-runs/l2_v1}"                          # L2 run dir (empty -> L1-only)
+PORT="${PORT:-5577}"
+cd "$(dirname "$0")"
+L2ARG=""; [ -n "$RUN2" ] && L2ARG="--l2run $RUN2"
+exec "$VENV/bin/python" infer_server.py --run "$RUN" $L2ARG --host 127.0.0.1 --port "$PORT"