feat(tests): revamp tests and remove legacy Dockerfiles: adopt JSON/consensus workflows, switch MiniCPM model, and delete deprecated Docker/test variants

image_support_files/paddleocr_vl_entrypoint.sh

@@ -1,19 +0,0 @@
-#!/bin/bash
-set -e
-
-echo "==================================="
-echo "PaddleOCR-VL Server (CPU)"
-echo "==================================="
-
-HOST="${SERVER_HOST:-0.0.0.0}"
-PORT="${SERVER_PORT:-8000}"
-
-echo "Host: ${HOST}"
-echo "Port: ${PORT}"
-echo "Device: CPU (no GPU)"
-echo ""
-
-echo "Starting PaddleOCR-VL CPU server..."
-echo "==================================="
-
-exec python /app/paddleocr_vl_server.py

image_support_files/paddleocr_vl_full_entrypoint.sh

@@ -1,12 +0,0 @@
-#!/bin/bash
-set -e
-
-echo "Starting PaddleOCR-VL Full Pipeline Server (Transformers backend)..."
-
-# Environment
-SERVER_PORT=${SERVER_PORT:-8000}
-SERVER_HOST=${SERVER_HOST:-0.0.0.0}
-
-# Start our API server directly (no vLLM - uses local transformers inference)
-echo "Starting API server on port $SERVER_PORT..."
-exec python /app/server.py