Spark Compute Isolation enabled (Azure#10726)

* Spark Compute Isolation enabled 

Using this flag the user can enabled the Spark compute in isolation

* Fix code style issue

Co-authored-by: Dongwei Wang <dongwwa@microsoft.com>

specification/synapse/resource-manager/Microsoft.Synapse/preview/2019-06-01-preview/bigDataPool.json

@@ -406,6 +406,10 @@
           "$ref": "#/definitions/AutoPauseProperties",
           "description": "Auto-pausing properties"
         },
+        "isComputeIsolationEnabled": {
+          "type": "boolean",
+          "description": "Whether compute isolation is required or not."
+        },
         "sparkEventsFolder": {
           "type": "string",
           "description": "The Spark events folder"
@@ -432,7 +436,9 @@
             "None",
             "Small",
             "Medium",
-            "Large"
+            "Large",
+            "XLarge",
+            "XXLarge"
           ],
           "type": "string",
           "description": "The level of compute power that each node in the Big Data pool has.",