[SPARK-28281][SQL][PYTHON][TESTS] Convert and port 'having.sql' into UDF test base

## What changes were proposed in this pull request?

This PR adds some tests converted from having.sql to test UDFs following the combination guide in [SPARK-27921](url)
<details><summary>Diff comparing to 'having.sql'</summary>
<p>

```diff
diff --git a/sql/core/src/test/resources/sql-tests/results/having.sql.out b/sql/core/src/test/resources/sql-tests/results/udf/udf-having.sql.out
index d87ee52..7cea2e5 100644
--- a/sql/core/src/test/resources/sql-tests/results/having.sql.out
+++ b/sql/core/src/test/resources/sql-tests/results/udf/udf-having.sql.out
 -16,34 +16,34  struct<>

 -- !query 1
-SELECT k, sum(v) FROM hav GROUP BY k HAVING sum(v) > 2
+SELECT udf(k) AS k, udf(sum(v)) FROM hav GROUP BY k HAVING udf(sum(v)) > 2
 -- !query 1 schema
-struct<k:string,sum(v):bigint>
+struct<k:string,udf(sum(cast(v as bigint))):string>
 -- !query 1 output
 one    6
 three  3

 -- !query 2
-SELECT count(k) FROM hav GROUP BY v + 1 HAVING v + 1 = 2
+SELECT udf(count(udf(k))) FROM hav GROUP BY v + 1 HAVING v + 1 = udf(2)
 -- !query 2 schema
-struct<count(k):bigint>
+struct<udf(count(udf(k))):string>
 -- !query 2 output
 1

 -- !query 3
-SELECT MIN(t.v) FROM (SELECT * FROM hav WHERE v > 0) t HAVING(COUNT(1) > 0)
+SELECT udf(MIN(t.v)) FROM (SELECT * FROM hav WHERE v > 0) t HAVING(udf(COUNT(udf(1))) > 0)
 -- !query 3 schema
-struct<min(v):int>
+struct<udf(min(v)):string>
 -- !query 3 output
 1

 -- !query 4
-SELECT a + b FROM VALUES (1L, 2), (3L, 4) AS T(a, b) GROUP BY a + b HAVING a + b > 1
+SELECT udf(a + b) FROM VALUES (1L, 2), (3L, 4) AS T(a, b) GROUP BY a + b HAVING a + b > udf(1)
 -- !query 4 schema
-struct<(a + CAST(b AS BIGINT)):bigint>
+struct<udf((a + cast(b as bigint))):string>
 -- !query 4 output
 3
 7

```

</p>
</details>

## How was this patch tested?

Tested as guided in SPARK-27921.

Closes apache#25093 from huaxingao/spark-28281.

Authored-by: Huaxin Gao <[email protected]>
Signed-off-by: HyukjinKwon <[email protected]>

Author: huaxingao

sql/core/src/test/resources/sql-tests/inputs/udf/udf-having.sql

@@ -0,0 +1,22 @@
+-- This test file was converted from having.sql.
+-- Note that currently registered UDF returns a string. So there are some differences, for instance
+-- in string cast within UDF in Scala and Python.
+
+create temporary view hav as select * from values
+  ("one", 1),
+  ("two", 2),
+  ("three", 3),
+  ("one", 5)
+  as hav(k, v);
+
+-- having clause
+SELECT udf(k) AS k, udf(sum(v)) FROM hav GROUP BY k HAVING udf(sum(v)) > 2;
+
+-- having condition contains grouping column
+SELECT udf(count(udf(k))) FROM hav GROUP BY v + 1 HAVING v + 1 = udf(2);
+
+-- SPARK-11032: resolve having correctly
+SELECT udf(MIN(t.v)) FROM (SELECT * FROM hav WHERE v > 0) t HAVING(udf(COUNT(udf(1))) > 0);
+
+-- SPARK-20329: make sure we handle timezones correctly
+SELECT udf(a + b) FROM VALUES (1L, 2), (3L, 4) AS T(a, b) GROUP BY a + b HAVING a + b > udf(1);

sql/core/src/test/resources/sql-tests/results/udf/udf-having.sql.out

@@ -0,0 +1,49 @@
+-- Automatically generated by SQLQueryTestSuite
+-- Number of queries: 5
+
+
+-- !query 0
+create temporary view hav as select * from values
+  ("one", 1),
+  ("two", 2),
+  ("three", 3),
+  ("one", 5)
+  as hav(k, v)
+-- !query 0 schema
+struct<>
+-- !query 0 output
+
+
+
+-- !query 1
+SELECT udf(k) AS k, udf(sum(v)) FROM hav GROUP BY k HAVING udf(sum(v)) > 2
+-- !query 1 schema
+struct<k:string,udf(sum(cast(v as bigint))):string>
+-- !query 1 output
+one	6
+three	3
+
+
+-- !query 2
+SELECT udf(count(udf(k))) FROM hav GROUP BY v + 1 HAVING v + 1 = udf(2)
+-- !query 2 schema
+struct<udf(count(udf(k))):string>
+-- !query 2 output
+1
+
+
+-- !query 3
+SELECT udf(MIN(t.v)) FROM (SELECT * FROM hav WHERE v > 0) t HAVING(udf(COUNT(udf(1))) > 0)
+-- !query 3 schema
+struct<udf(min(v)):string>
+-- !query 3 output
+1
+
+
+-- !query 4
+SELECT udf(a + b) FROM VALUES (1L, 2), (3L, 4) AS T(a, b) GROUP BY a + b HAVING a + b > udf(1)
+-- !query 4 schema
+struct<udf((a + cast(b as bigint))):string>
+-- !query 4 output
+3
+7