.Net: Use AsynEnumerable search interface for azure ai search. (#9090)

### Motivation and Context

AzureAISearch has two ways of accessing search results.

### Description

- Switching to IAsyncEnumerable interface to improve performance.
- Switch to AzureOpenAI for integration test embeddings to remove need
for key management.
- Fix small bug in powershell module to create azure ai search test
infra

#9077

### Contribution Checklist

<!-- Before submitting this PR, please make sure: -->

- [x] The code builds clean without any errors or warnings
- [x] The PR follows the [SK Contribution
Guidelines](https://github.com/microsoft/semantic-kernel/blob/main/CONTRIBUTING.md)
and the [pre-submission formatting
script](https://github.com/microsoft/semantic-kernel/blob/main/CONTRIBUTING.md#development-scripts)
raises no violations
- [x] All unit tests pass, and I have added new tests where possible
- [x] I didn't break anyone 😄

dotnet/src/Connectors/Connectors.Memory.AzureAISearch/AzureAISearchVectorStoreRecordCollection.cs

@@ -446,7 +446,7 @@ public Task<VectorSearchResults<TRecord>> VectorizableTextSearchAsync(string sea
     /// <param name="searchOptions">The options controlling the behavior of the search operation.</param>
     /// <param name="includeVectors">A value indicating whether to include vectors in the result or not.</param>
     /// <param name="cancellationToken">The <see cref="CancellationToken"/> to monitor for cancellation requests. The default is <see cref="CancellationToken.None"/>.</param>
-    /// <returns></returns>
+    /// <returns>The mapped search results.</returns>
     private async Task<VectorSearchResults<TRecord>> SearchAndMapToDataModelAsync(
         string? searchText,
         SearchOptions searchOptions,
@@ -462,23 +462,14 @@ private async Task<VectorSearchResults<TRecord>> SearchAndMapToDataModelAsync(
                 OperationName,
                 () => this._searchClient.SearchAsync<JsonObject>(searchText, searchOptions, cancellationToken)).ConfigureAwait(false);
 
-            var mappedJsonObjectResults = jsonObjectResults.Value.GetResults().Select(result =>
-            {
-                var document = VectorStoreErrorHandler.RunModelConversion(
-                    DatabaseName,
-                    this._collectionName,
-                    OperationName,
-                    () => this._options.JsonObjectCustomMapper!.MapFromStorageToDataModel(result.Document, new() { IncludeVectors = includeVectors }));
-                return new VectorSearchResult<TRecord>(document, result.Score);
-            });
-
-            return new VectorSearchResults<TRecord>(mappedJsonObjectResults.ToAsyncEnumerable()) { TotalCount = jsonObjectResults.Value.TotalCount };
+            var mappedJsonObjectResults = this.MapSearchResultsAsync(jsonObjectResults.Value.GetResultsAsync(), OperationName, includeVectors);
+            return new VectorSearchResults<TRecord>(mappedJsonObjectResults) { TotalCount = jsonObjectResults.Value.TotalCount };
         }
 
         // Execute search and map using the built in Azure AI Search mapper.
         Response<SearchResults<TRecord>> results = await this.RunOperationAsync(OperationName, () => this._searchClient.SearchAsync<TRecord>(searchText, searchOptions, cancellationToken)).ConfigureAwait(false);
-        var mappedResults = results.Value.GetResults().Select(result => new VectorSearchResult<TRecord>(result.Document, result.Score));
-        return new VectorSearchResults<TRecord>(mappedResults.ToAsyncEnumerable()) { TotalCount = results.Value.TotalCount };
+        var mappedResults = this.MapSearchResultsAsync(results.Value.GetResultsAsync());
+        return new VectorSearchResults<TRecord>(mappedResults) { TotalCount = results.Value.TotalCount };
     }
 
     /// <summary>
@@ -515,6 +506,39 @@ private Task<Response<IndexDocumentsResult>> MapToStorageModelAndUploadDocumentA
             () => this._searchClient.UploadDocumentsAsync<TRecord>(records, innerOptions, cancellationToken));
     }
 
+    /// <summary>
+    /// Map the search results from <see cref="SearchResult{JsonObject}"/> to <see cref="VectorSearchResults{TRecord}"/> objects using the configured mapper type.
+    /// </summary>
+    /// <param name="results">The search results to map.</param>
+    /// <param name="operationName">The name of the current operation for telemetry purposes.</param>
+    /// <param name="includeVectors">A value indicating whether to include vectors in the resultset or not.</param>
+    /// <returns>The mapped results.</returns>
+    private async IAsyncEnumerable<VectorSearchResult<TRecord>> MapSearchResultsAsync(IAsyncEnumerable<SearchResult<JsonObject>> results, string operationName, bool includeVectors)
+    {
+        await foreach (var result in results.ConfigureAwait(false))
+        {
+            var document = VectorStoreErrorHandler.RunModelConversion(
+                DatabaseName,
+                this._collectionName,
+                operationName,
+                () => this._options.JsonObjectCustomMapper!.MapFromStorageToDataModel(result.Document, new() { IncludeVectors = includeVectors }));
+            yield return new VectorSearchResult<TRecord>(document, result.Score);
+        }
+    }
+
+    /// <summary>
+    /// Map the search results from <see cref="SearchResult{TRecord}"/> to <see cref="VectorSearchResults{TRecord}"/> objects.
+    /// </summary>
+    /// <param name="results">The search results to map.</param>
+    /// <returns>The mapped results.</returns>
+    private async IAsyncEnumerable<VectorSearchResult<TRecord>> MapSearchResultsAsync(IAsyncEnumerable<SearchResult<TRecord>> results)
+    {
+        await foreach (var result in results.ConfigureAwait(false))
+        {
+            yield return new VectorSearchResult<TRecord>(result.Document, result.Score);
+        }
+    }
+
     /// <summary>
     /// Convert the public <see cref="GetRecordOptions"/> options model to the Azure AI Search <see cref="GetDocumentOptions"/> options model.
     /// </summary>

dotnet/src/IntegrationTests/Connectors/Memory/AzureAISearch/AzureAISearchTextSearchTests.cs

@@ -2,9 +2,10 @@
 
 using System;
 using System.Threading.Tasks;
+using Azure.Identity;
 using Microsoft.Extensions.Configuration;
 using Microsoft.SemanticKernel.Connectors.AzureAISearch;
-using Microsoft.SemanticKernel.Connectors.OpenAI;
+using Microsoft.SemanticKernel.Connectors.AzureOpenAI;
 using Microsoft.SemanticKernel.Data;
 using SemanticKernel.IntegrationTests.Data;
 using SemanticKernel.IntegrationTests.TestSettings;
@@ -69,11 +70,14 @@ public override Task<ITextSearch> CreateTextSearchAsync()
     {
         if (this.VectorStore is null)
         {
-            OpenAIConfiguration? openAIConfiguration = this.Configuration.GetSection("OpenAIEmbeddings").Get<OpenAIConfiguration>();
-            Assert.NotNull(openAIConfiguration);
-            Assert.NotEmpty(openAIConfiguration.ModelId);
-            Assert.NotEmpty(openAIConfiguration.ApiKey);
-            this.EmbeddingGenerator = new OpenAITextEmbeddingGenerationService(openAIConfiguration.ModelId, openAIConfiguration.ApiKey);
+            AzureOpenAIConfiguration? azureOpenAIConfiguration = this.Configuration.GetSection("AzureOpenAIEmbeddings").Get<AzureOpenAIConfiguration>();
+            Assert.NotNull(azureOpenAIConfiguration);
+            Assert.NotEmpty(azureOpenAIConfiguration.DeploymentName);
+            Assert.NotEmpty(azureOpenAIConfiguration.Endpoint);
+            this.EmbeddingGenerator = new AzureOpenAITextEmbeddingGenerationService(
+                azureOpenAIConfiguration.DeploymentName,
+                azureOpenAIConfiguration.Endpoint,
+                new AzureCliCredential());
 
             this.VectorStore = new AzureAISearchVectorStore(fixture.SearchIndexClient);
         }

dotnet/src/IntegrationTests/TestSettings/Memory/AzureAISearchSetup.psm1

@@ -32,7 +32,7 @@ function New-AzureAISearchIntegrationInfra($overrideResourceGroup = $resourceGro
     }
 
     # Create the ai search service if it doesn't exist.
-    $service = Get-AzSearchService -ResourceGroupName $resourceGroup -Name $aiSearchResourceName
+    $service = Get-AzSearchService -ResourceGroupName $overrideResourceGroup -Name $overrideAISearchResourceName
     if (-not $service) {
         Write-Host "Service does not exist, creating '$overrideAISearchResourceName' ..."
         New-AzSearchService -ResourceGroupName $overrideResourceGroup -Name $overrideAISearchResourceName -Sku "Basic" -Location "North Europe" -PartitionCount 1 -ReplicaCount 1 -HostingMode Default