Fix: fix sagemaker-serve tests (#5407)

* fix: Fix the recipe selection for multiple recipe scenario

* fix: Fix the recipe selection for multiple recipe scenario

* fix: Hyperparameter issue fixes, validate s3 output path,additional unit tests

* Fix: Add validation to bedrock reward models

* Fix: Add validation to bedrock reward models

* Fix: Add allow list for bedrock eval models

* Fix: Add allow list for bedrock eval models

* Fix: Bug fixes for s3 path validation, mlflow app creation

* Fix: Update Legal verbiage, and allowed reward model ids based on region

* Fix: Update model_package_group_name to model_package_group in all trianers to maintain consistency

* Fix: fix sagemaker-serve tests

---------

Co-authored-by: Roja Reddy Sareddy <[email protected]>

Author: rsareddy0329

sagemaker-serve/tests/integ/test_model_customization_deployment.py

@@ -255,7 +255,7 @@ def test_sft_trainer_build(self, training_job_name):
             model="meta-textgeneration-llama-3-2-1b-instruct",
             training_dataset="s3://dummy/data.jsonl",
             accept_eula=True,
-            model_package_group_name="test-group"
+            model_package_group="test-group"
         )
         trainer._latest_training_job = training_job
 
@@ -282,7 +282,7 @@ def test_dpo_trainer_build(self, training_job_name):
                 model="meta-textgeneration-llama-3-2-1b-instruct",
                 training_dataset="s3://dummy/data.jsonl",
                 accept_eula=True,
-                model_package_group_name="test-group"
+                model_package_group="test-group"
             )
         trainer._latest_training_job = training_job