fix(job_train): drop flash-attn install to avoid build-time OOM (R3 mitigation)

scripts/job_train.sh

@@ -29,7 +29,9 @@ pip install -q vllm==0.18.0
 # Optional transitive deps often required by quantized loaders / datasets stack
 pip install -q safetensors sentencepiece einops scipy xxhash protobuf pyyaml fsspec aiohttp dill multiprocess pyarrow requests filelock packaging tokenizers regex tqdm
 
-pip install flash-attn --no-build-isolation -q || echo "flash-attn skipped"
+# flash-attn removed: source build is the largest single memory spike during install.
+# vLLM + transformers fall back to PyTorch SDPA without it (small throughput cost).
+echo "flash-attn intentionally skipped to avoid build-time OOM"
 
 nohup uvicorn env.server.app:app --host 0.0.0.0 --port 8001 > env.log 2>&1 &
 sleep 10