[evaluation] Remove _InternalRiskCategory from type hint (#42308)

* chore: Update date on changelog

* fix(typing): Remove non-public enum from type hint

         apiview feedback

Author: kdestin

sdk/evaluation/azure-ai-evaluation/CHANGELOG.md

@@ -1,6 +1,6 @@
 # Release History
 
-## 1.10.0 (2025-07-29)
+## 1.10.0 (2025-07-31)
 
 ### Breaking Changes
 

sdk/evaluation/azure-ai-evaluation/azure/ai/evaluation/red_team/_red_team.py

@@ -280,7 +280,7 @@ def __init__(
         application_scenario: Optional[str] = None,
         custom_attack_seed_prompts: Optional[str] = None,
         output_dir=".",
-        attack_success_thresholds: Optional[Dict[Union[RiskCategory, _InternalRiskCategory], int]] = None,
+        attack_success_thresholds: Optional[Dict[RiskCategory, int]] = None,
     ):
         """Initialize a new Red Team agent for AI model evaluation.
 
@@ -307,7 +307,7 @@ def __init__(
             Should be a dictionary mapping risk categories (RiskCategory enum values) to threshold values,
             or None to use default binary evaluation (evaluation results determine success).
             When using thresholds, scores >= threshold are considered successful attacks.
-        :type attack_success_thresholds: Optional[Dict[Union[RiskCategory, _InternalRiskCategory], int]]
+        :type attack_success_thresholds: Optional[Dict[RiskCategory, int]]
         """
 
         self.azure_ai_project = validate_azure_ai_project(azure_ai_project)