Enable model cache for Rocm docker compose test. (#1614)

Signed-off-by: ZePan110 <ze.pan@intel.com>

Translation/docker_compose/amd/gpu/rocm/compose.yaml

@@ -8,7 +8,7 @@ services:
     ports:
       - "${TRANSLATION_TGI_SERVICE_PORT:-8008}:80"
     volumes:
-      - "/var/lib/GenAI/translation/data:/data"
+      - "${MODEL_CACHE:-/var/lib/GenAI/translation/data}:/data"
     shm_size: 8g
     environment:
       no_proxy: ${no_proxy}