Add tests and docs for regexp() and regexp_like()

Signed-off-by: Andy Grove <andygrove@nvidia.com>

docs/compatibility.md

@@ -271,10 +271,15 @@ The RAPIDS Accelerator for Apache Spark currently supports string literal matche
 matches for the `regexp_replace` function and will fall back to CPU if a regular expression pattern 
 is provided.
 
-### RLike
+### RLike, regexp(), and regexp_like()
 
-The GPU implementation of `RLike` has the following known issues where behavior is not consistent with Apache Spark and
-this expression is disabled by default. It can be enabled setting `spark.rapids.sql.expression.RLike=true`.
+The GPU implementation of `RLike` is disabled by default. It can be enabled by setting 
+`spark.rapids.sql.expression.RLike=true`.
+
+Apache Spark 3.2.0 introduces the `regexp` and `regexp_like` functions which are equivalent to the `RLike` 
+expression and are also supported on GPU when `RLike` is enabled on the GPU.
+
+`RLike` is disabled by default due to the following known issues where behavior is not consistent with Apache Spark.
 
 - `$` does not match the end of string if the string ends with a line-terminator 
   ([cuDF issue #9620](https://github.com/rapidsai/cudf/issues/9620))

integration_tests/src/main/python/string_test.py

@@ -469,7 +469,29 @@ def test_like_complex_escape():
                 'a like "\\%SystemDrive\\%\\\\\\\\Users%"',
                 'a like "_oo"'),
             conf={'spark.sql.parser.escapedStringLiterals': 'true'})
+
+@pytest.mark.skipif(not is_before_spark_320() or is_databricks91_or_later(), reason='regexp is synonym for RLike starting in Spark 3.2.0')
+def test_regexp():
+    gen = mk_str_gen('[abcd]{1,3}')
+    assert_gpu_and_cpu_are_equal_collect(
+            lambda spark: unary_op_df(spark, gen).selectExpr(
+                'regexp(a, "a{2}")',
+                'regexp(a, "a{1,3}")',
+                'regexp(a, "a{1,}")',
+                'regexp(a, "a[bc]d")'),
+            conf={'spark.rapids.sql.expression.RLike': 'true'})
 
+@pytest.mark.skipif(not is_before_spark_320() or is_databricks91_or_later(), reason='regexp_like is synonym for RLike starting in Spark 3.2.0')
+def test_regexp_like():
+    gen = mk_str_gen('[abcd]{1,3}')
+    assert_gpu_and_cpu_are_equal_collect(
+            lambda spark: unary_op_df(spark, gen).selectExpr(
+                'regexp_like(a, "a{2}")',
+                'regexp_like(a, "a{1,3}")',
+                'regexp_like(a, "a{1,}")',
+                'regexp_like(a, "a[bc]d")'),
+            conf={'spark.rapids.sql.expression.RLike': 'true'})
+
 def test_rlike():
     gen = mk_str_gen('[abcd]{1,3}')
     assert_gpu_and_cpu_are_equal_collect(