Spark 3.5: Fix Javadoc in ColumnarBatchUtil

[huaxingao]

#12053

[huaxingao]

cc @aokolnychyi @szehon-ho @dramaticlly

[aokolnychyi]

I just realized the example actually doesn't fit anymore as we only compute the mapping, not the isDeleted array. Also, what if we use <pre> blocks to avoid disabling Spotless?

/**
 * Builds a row ID mapping inside a batch to skip deleted rows.
 *
 * <pre>
 * Initial state
 * Data values: [v0, v1, v2, v3, v4, v5, v6, v7]
 * Row ID mapping: [0, 1, 2, 3, 4, 5, 6, 7]
 *
 * Apply position deletes
 * Position deletes: 2, 6
 * Row ID mapping: [0, 1, 3, 4, 5, 7, -, -] (6 live records)
 *
 * Apply equality deletes
 * Equality deletes: v1, v2, v3
 * Row ID mapping: [0, 4, 5, 7, -, -, -, -] (4 live records)
 * </pre>
 *
 * @param columnVectors the array of column vectors for the batch
 * @param deletes the delete filter containing delete information
 * @param rowStartPosInBatch the starting position of the row in the batch
 * @param batchSize the size of the batch
 * @return the mapping array and the number of live rows, or {@code null} if nothing is deleted
*/

I think this example would be more descriptive as it is clear what the data values are and which records are removed by equality deletes.

[aokolnychyi]

Let's also give a similar example in the buildIsDeleted method below.

[huaxingao]

Fixed. Thanks!