update trl to 0.17.0 (#2560)

* update trl to 0.17.0

* grpo + vllm no longer supported with 2.5.1 due to vllm constraints

* disable VLLM_USE_V1 for ci

* imporve handle killing off of multiprocessing vllm service

* debug why this doesn't run in CI

* increase vllm wait time

* increase timeout to 5min

* upgrade to vllm 0.8.4

* dump out the vllm log for debugging

* use debug logging

* increase vllm start timeout

* use NVL instead

* disable torch compile cache

* revert some commented checks now that grpo tests are fixed

* increase vllm timeoout back to 5min

.github/workflows/multi-gpu-e2e.yml

@@ -43,7 +43,7 @@ jobs:
             cuda_version: 12.4.1
             python_version: "3.11"
             pytorch: 2.5.1
-            axolotl_extras: vllm
+            axolotl_extras:
             num_gpus: 2
             nightly_build: "true"
           - cuda: 126