Update GPT-5.4 evals to GPT-6.1 Sol (#50874)

Co-authored-by: helen229 <gaoh@microsoft.com>

Author: azure-sdk-automation[bot]

.github/skills/azsdk-common-apiview-feedback-resolution/evals/eval.yaml

@@ -11,7 +11,7 @@ tags:
 config:
   runs: 1
   timeout: "90s"
-  model: gpt-5.4
+  model: gpt-6.1-sol
   executor: copilot-sdk
 
 scoring:

.github/skills/azsdk-common-apiview-feedback-resolution/evals/trigger.eval.yaml

@@ -11,7 +11,7 @@ tags:
 defaults:
   runs: 1
   timeout: "90s"
-  model: gpt-5.4
+  model: gpt-6.1-sol
   executor: copilot-sdk
 
 scoring:

.github/skills/azsdk-common-generate-sdk-locally/evals/eval.yaml

@@ -11,7 +11,7 @@ tags:
 config:
   runs: 1
   timeout: "90s"
-  model: gpt-5.4
+  model: gpt-6.1-sol
   executor: copilot-sdk
 
 scoring:

.github/skills/azsdk-common-pipeline-analysis/evals/trigger.eval.yaml

@@ -11,7 +11,7 @@ tags:
 defaults:
   runs: 1
   timeout: "90s"
-  model: gpt-5.4
+  model: gpt-6.1-sol
   executor: copilot-sdk
 
 scoring:

.github/skills/azsdk-common-pipeline-fixer/evals/trigger.eval.yaml

@@ -11,11 +11,11 @@ tags:
 defaults:
   runs: 1
   # Stimuli with a resolvable build reference (URL or build ID) make the agent
-  # run the full analyze→fix→verify workflow (~90-180s under gpt-5.4),
+  # run the full analyze→fix→verify workflow (~90-180s under gpt-6.1-sol),
   # so the per-stimulus cap matches the Vally scenario evals (5m). Non-build-ref
   # routing prompts still complete in ~10-20s.
   timeout: "5m"
-  model: gpt-5.4
+  model: gpt-6.1-sol
   executor: copilot-sdk
 
 scoring:

.github/skills/azsdk-common-prepare-release-plan/evals/eval.yaml

@@ -11,7 +11,7 @@ tags:
 config:
   runs: 1
   timeout: "90s"
-  model: gpt-5.4
+  model: gpt-6.1-sol
   executor: copilot-sdk
 
 scoring:

.github/skills/azsdk-common-prepare-release-plan/evals/schedule-risk.eval.yaml

@@ -11,7 +11,7 @@ tags:
 defaults:
   runs: 1
   timeout: "120s"
-  model: gpt-5.4
+  model: gpt-6.1-sol
   executor: copilot-sdk
 
 scoring:

.github/skills/azsdk-common-prepare-release-plan/evals/trigger.eval.yaml

@@ -11,7 +11,7 @@ tags:
 defaults:
   runs: 1
   timeout: "90s"
-  model: gpt-5.4
+  model: gpt-6.1-sol
   executor: copilot-sdk
 
 scoring:

.github/skills/azsdk-common-sdk-release/evals/eval.yaml

@@ -11,7 +11,7 @@ tags:
 config:
   runs: 1
   timeout: "90s"
-  model: gpt-5.4
+  model: gpt-6.1-sol
   executor: copilot-sdk
 
 scoring:

.github/skills/azsdk-common-sdk-release/evals/trigger.eval.yaml

@@ -11,7 +11,7 @@ tags:
 defaults:
   runs: 1
   timeout: "90s"
-  model: gpt-5.4
+  model: gpt-6.1-sol
   executor: copilot-sdk
 
 scoring: