fix: remove hardcoded PostgreSQL credentials from docker-compose.yml

.env.example

@@ -12,6 +12,12 @@ SECUSCAN_BIND_PORT=8000
 # SECUSCAN_REDIS_URL=redis://127.0.0.1:6379/0
 # SECUSCAN_CACHE_TTL_SECONDS=30
 
+# PostgreSQL — REQUIRED when using docker-compose (not for SQLite dev)
+# Generate a strong password with: python -c "import secrets; print(secrets.token_urlsafe(32))"
+# SECUSCAN_POSTGRES_USER=secuscan
+# SECUSCAN_POSTGRES_DB=secuscan
+# SECUSCAN_POSTGRES_PASSWORD=
+
 # Docker Support
 SECUSCAN_DOCKER_ENABLED=false
 # Docker sandbox network (auto-created if absent; ICC disabled for isolation)

docker-compose.yml

@@ -3,15 +3,15 @@ services:
   postgres:
     image: postgres:16-alpine
     environment:
-      POSTGRES_DB: secuscan
-      POSTGRES_USER: secuscan
-      POSTGRES_PASSWORD: secuscan
+      POSTGRES_DB: ${SECUSCAN_POSTGRES_DB:-secuscan}
+      POSTGRES_USER: ${SECUSCAN_POSTGRES_USER:-secuscan}
+      POSTGRES_PASSWORD: ${SECUSCAN_POSTGRES_PASSWORD:?Set SECUSCAN_POSTGRES_PASSWORD to a strong password}
     ports:
       - "127.0.0.1:5432:5432"
     volumes:
       - postgres_data:/var/lib/postgresql/data
     healthcheck:
-      test: ["CMD-SHELL", "pg_isready -U secuscan -d secuscan"]
+      test: ["CMD-SHELL", "pg_isready -U ${SECUSCAN_POSTGRES_USER:-secuscan} -d ${SECUSCAN_POSTGRES_DB:-secuscan}"]
       interval: 10s
       timeout: 5s
       retries: 5
@@ -43,7 +43,7 @@ services:
     environment:
       - SECUSCAN_BIND_ADDRESS=0.0.0.0   # fix: was 127.0.0.1, unreachable inside Docker
       - SECUSCAN_BIND_PORT=8081
-      - SECUSCAN_POSTGRES_DSN=postgresql://secuscan:secuscan@postgres:5432/secuscan
+      - SECUSCAN_POSTGRES_DSN=postgresql://${SECUSCAN_POSTGRES_USER:-secuscan}:${SECUSCAN_POSTGRES_PASSWORD:?}@postgres:5432/${SECUSCAN_POSTGRES_DB:-secuscan}
       - SECUSCAN_REDIS_URL=redis://redis:6379/0
       - SECUSCAN_PLUGINS_DIR=/app/plugins
     depends_on: