Test fixes: place n5 in HOME/Desktop/test.n5, add spark.driver.bindAd…

…dress setting, and create n5 dataset before rendering it. To run in IntelliJ, edit run configuration classpath to include HOME/.m2/repository/org/apache/hadoop/hadoop-client-runtime/3.3.4/hadoop-client-runtime-3.3.4.jar and select "Add dependencies with "provided" scope to classpath".

render-ws-spark-client/src/test/java/org/janelia/render/client/spark/n5/N5ClientTest.java

@@ -274,13 +274,14 @@ public static void main(final String[] args) {
         parameters.stack = "s001_m239_align_no35";
         parameters.tileWidth = 2048;
         parameters.tileHeight = 2048;
-        parameters.n5Path = "test.n5";
+        parameters.n5Path = System.getenv("HOME") + "/Desktop/test.n5";
         parameters.n5Dataset = "/output";
 
         final N5Client n5Client = new N5Client(parameters);
 
         final SparkConf sparkConf = new SparkConf()
                 .setMaster("local[4]") // run spark locally with 4 threads
+                .set("spark.driver.bindAddress", "127.0.0.1")
                 .setAppName("test");
         final JavaSparkContext sparkContext = new JavaSparkContext(sparkConf);
 
@@ -300,6 +301,14 @@ public static void main(final String[] args) {
         };
         final ImageProcessorCacheSpec cacheSpec = N5Client.buildImageProcessorCacheSpec();
 
+        try (final N5Writer n5 = new N5FSWriter(parameters.n5Path)) {
+            n5.createDataset(fullScaleDatasetName,
+                             dimensions,
+                             blockSize,
+                             DataType.UINT8,
+                             new GzipCompression());
+        }
+
         n5Client.renderStack(sparkContext,
                              blockSize,
                              fullScaleDatasetName,