[SPARKR][MINOR] Fix example of spark.naiveBayes

## What changes were proposed in this pull request?

The original example doesn't work because the features are not categorical. This PR fixes this by changing to another dataset.

## How was this patch tested?

Manual test.

Author: Junyang Qian <junyangq@databricks.com>

Closes #14820 from junyangq/SPARK-FixNaiveBayes.

R/pkg/R/mllib.R

@@ -747,10 +747,11 @@ setMethod("summary", signature(object = "MultilayerPerceptronClassificationModel
 #' @export
 #' @examples
 #' \dontrun{
-#' df <- createDataFrame(infert)
+#' data <- as.data.frame(UCBAdmissions)
+#' df <- createDataFrame(data)
 #'
 #' # fit a Bernoulli naive Bayes model
-#' model <- spark.naiveBayes(df, education ~ ., smoothing = 0)
+#' model <- spark.naiveBayes(df, Admit ~ Gender + Dept, smoothing = 0)
 #'
 #' # get the summary of the model
 #' summary(model)