fix(pytorch): reduce 2.13 GPU max_jobs to 4 to fit wheel-export in RAM

The wheel-cache upload path in
scripts/ci/build/pytorch_runtime/lib/upload_wheels.sh runs
`docker buildx build --target wheel-export`, which re-executes the
entire Dockerfile to reach a scratch stage that COPYs the built wheels
out of the flash-attn and TE builder stages. With MAX_JOBS=8, the
concurrent flash-attn 2.8.3 + transformer-engine 2.16.1 source builds
during that re-execution exceed the CI runner cgroup memory limit and
get OOM-killed (SIGKILL / exit code 137, PR #6365 attempt 2).

Halving MAX_JOBS to 4 roughly halves peak compile RAM, keeping the
re-execution under the runner ceiling at the cost of a slower flash-attn
/ TE compile step. CPU configs unaffected.

Author: Bhanu Teja Goshikonda

.github/config/image/pytorch/2.13-ec2-cuda.yml

@@ -26,7 +26,7 @@ build:
   nccl_version: "2.30.7-1"
   efa_version: "1.49.0"
   gdrcopy_version: "2.6"
-  max_jobs: "8"
+  max_jobs: "4"
   dlc_major_version: "1"
   dlc_minor_version: "0"
 

.github/config/image/pytorch/2.13-sagemaker-cuda.yml

@@ -27,7 +27,7 @@ build:
   nccl_version: "2.30.7-1"
   efa_version: "1.49.0"
   gdrcopy_version: "2.6"
-  max_jobs: "8"
+  max_jobs: "4"
   dlc_major_version: "1"
   dlc_minor_version: "0"