Fix parallel CP test port conflicts in L1 distributed test.sh

When NUM_GPUS >= 8, test.sh runs two parallel pytest sessions —
non-deterministic on GPUs 0-3 and deterministic on GPUs 4-7. Both
sessions launch their own torchrun (or pool worker) which picks
MASTER_PORT from the environment. Without explicit ports, both
inherit the same default (29500) and the second session fails with
EADDRINUSE on every torchrun spawn.

Assign distinct ports: 29500 for the non-deterministic session,
29501 for the deterministic session.

Picked up the OOM skip half of fdf32d5 separately during the
pool-batching rebase; this commit closes the remaining gap.

Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

Author: sudhakarsingh27

qa/L1_pytorch_distributed_unittest/test.sh

@@ -28,9 +28,9 @@ NUM_GPUS=$(python3 -c "import torch; print(torch.cuda.device_count())")
 echo "Detected $NUM_GPUS GPU(s)"
 if [ "$NUM_GPUS" -ge 8 ]; then
     echo "Running CP tests in parallel: non-deterministic on GPUs 0-3, deterministic on GPUs 4-7"
-    CUDA_VISIBLE_DEVICES=0,1,2,3 python3 -m pytest -v -s --junitxml=$XML_LOG_DIR/pytest_test_attention_with_cp.xml $TE_PATH/tests/pytorch/attention/test_attention_with_cp.py &
+    CUDA_VISIBLE_DEVICES=0,1,2,3 MASTER_PORT=29500 python3 -m pytest -v -s --junitxml=$XML_LOG_DIR/pytest_test_attention_with_cp.xml $TE_PATH/tests/pytorch/attention/test_attention_with_cp.py &
     PID_CP_NONDET=$!
-    CUDA_VISIBLE_DEVICES=4,5,6,7 NVTE_ALLOW_NONDETERMINISTIC_ALGO=0 python3 -m pytest -v -s --junitxml=$XML_LOG_DIR/pytest_test_attention_deterministic_with_cp.xml $TE_PATH/tests/pytorch/attention/test_attention_with_cp.py &
+    CUDA_VISIBLE_DEVICES=4,5,6,7 MASTER_PORT=29501 NVTE_ALLOW_NONDETERMINISTIC_ALGO=0 python3 -m pytest -v -s --junitxml=$XML_LOG_DIR/pytest_test_attention_deterministic_with_cp.xml $TE_PATH/tests/pytorch/attention/test_attention_with_cp.py &
     PID_CP_DET=$!
     wait $PID_CP_NONDET || test_fail "test_attention_with_cp.py"
     wait $PID_CP_DET || test_fail "NVTE_ALLOW_NONDETERMINISTIC_ALGO=0 test_attention_with_cp.py"