[Evaluation] Prepare azure-ai-evaluation 1.18.2 release

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Copilot-Session: a8d6b8cc-0ccd-4408-a420-22b63f844cb9

Author: slister1001

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

@@ -1,5 +1,18 @@
 # Release History
 
+## 1.18.2 (2026-07-21)
+
+### Features Added
+
+- Added support for admin-connected (BYO) judge models in prompty-based evaluators. Models configured
+  with `byo_model` and `project_endpoint` are routed through the Foundry project Responses API, which
+  resolves the connection and its authentication.
+
+### Bugs Fixed
+
+- Updated `TaskNavigationEfficiencyEvaluator` validation errors to name both accepted input aliases:
+  `response` / `actions` and `ground_truth` / `expected_actions`.
+
 ## 1.18.1 (2026-07-09)
 
 ### Bugs Fixed

sdk/evaluation/azure-ai-evaluation/azure/ai/evaluation/_version.py

@@ -3,4 +3,4 @@
 # ---------------------------------------------------------
 # represents upcoming version
 
-VERSION = "1.18.1"
+VERSION = "1.18.2"