[SPARK-24329][SQL] Test for skipping multi-space lines

## What changes were proposed in this pull request?

The PR is a continue of #21380 . It checks cases that are handled by the code:
https://github.com/apache/spark/blob/e3de6ab30d52890eb08578e55eb4a5d2b4e7aa35/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/csv/UnivocityParser.scala#L303-L304

Basically the code skips lines with one or many whitespaces, and lines with comments (see [filterCommentAndEmpty](https://github.com/apache/spark/blob/e3de6ab30d52890eb08578e55eb4a5d2b4e7aa35/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/csv/CSVUtils.scala#L47))

```scala
   iter.filter { line =>
      line.trim.nonEmpty && !line.startsWith(options.comment.toString)
    }
```

Closes #21380

## How was this patch tested?

Added a test for the case described above.

Author: Maxim Gekk <[email protected]>
Author: Maxim Gekk <[email protected]>

Closes #21394 from MaxGekk/test-for-multi-space-lines.

Author: MaxGekk

sql/core/src/test/resources/test-data/comments-whitespaces.csv

@@ -0,0 +1,8 @@
+# The file contains comments, whitespaces and empty lines
+colA
+# empty line
+
+# the line with a few whitespaces
+   
+# int value with leading and trailing whitespaces
+ "a" 

sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/csv/CSVSuite.scala

@@ -1368,4 +1368,19 @@ class CSVSuite extends QueryTest with SharedSQLContext with SQLTestUtils with Te
       checkAnswer(computed, expected)
     }
   }
+
+  test("SPARK-24329: skip lines with comments, and one or multiple whitespaces") {
+    val schema = new StructType().add("colA", StringType)
+    val ds = spark
+      .read
+      .schema(schema)
+      .option("multiLine", false)
+      .option("header", true)
+      .option("comment", "#")
+      .option("ignoreLeadingWhiteSpace", false)
+      .option("ignoreTrailingWhiteSpace", false)
+      .csv(testFile("test-data/comments-whitespaces.csv"))
+
+    checkAnswer(ds, Seq(Row(""" "a" """)))
+  }
 }