[SPARK-29237][SQL] Prevent real function names in expression example template

### What changes were proposed in this pull request?

In the PR, I propose to replace function names in some expression examples by `_FUNC_`, and add a test to check that `_FUNC_` always present in all examples.

### Why are the changes needed?
Binding of a function name to an expression is performed in `FunctionRegistry` which is single source of truth. Expression examples should avoid using function name directly because this can make the examples invalid in the future.

### Does this PR introduce any user-facing change?
No

### How was this patch tested?
Added new test to `SQLQuerySuite` which analyses expression example, and check presence of `_FUNC_`.

Closes #25924 from MaxGekk/fix-func-examples.

Authored-by: Maxim Gekk <max.gekk@gmail.com>
Signed-off-by: Dongjoon Hyun <dhyun@apple.com>

Author: MaxGekk

sql/catalyst/src/main/java/org/apache/spark/sql/catalyst/expressions/ExpressionInfo.java

@@ -17,6 +17,8 @@
 
 package org.apache.spark.sql.catalyst.expressions;
 
+import com.google.common.annotations.VisibleForTesting;
+
 /**
  * Expression information, will be used to describe a expression.
  */
@@ -56,6 +58,11 @@ public String getArguments() {
         return arguments;
     }
 
+    @VisibleForTesting
+    public String getOriginalExamples() {
+        return examples;
+    }
+
     public String getExamples() {
         return replaceFunctionName(examples);
     }

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/datetimeExpressions.scala

@@ -1097,7 +1097,7 @@ case class TimeAdd(start: Expression, interval: Expression, timeZoneId: Option[S
   usage = "_FUNC_(timestamp, timezone) - Given a timestamp like '2017-07-14 02:40:00.0', interprets it as a time in UTC, and renders that time as a timestamp in the given time zone. For example, 'GMT+1' would yield '2017-07-14 03:40:00.0'.",
   examples = """
     Examples:
-      > SELECT from_utc_timestamp('2016-08-31', 'Asia/Seoul');
+      > SELECT _FUNC_('2016-08-31', 'Asia/Seoul');
        2016-08-31 09:00:00
   """,
   since = "1.5.0",

sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala

@@ -115,6 +115,30 @@ class SQLQuerySuite extends QueryTest with SharedSparkSession {
     }
   }
 
+  test("using _FUNC_ instead of function names in examples") {
+    val exampleRe = "(>.*;)".r
+    val ignoreSet = Set(
+      // Examples for CaseWhen show simpler syntax:
+      // `CASE WHEN ... THEN ... WHEN ... THEN ... END`
+      "org.apache.spark.sql.catalyst.expressions.CaseWhen",
+      // _FUNC_ is replaced by `locate` but `locate(... IN ...)` is not supported
+      "org.apache.spark.sql.catalyst.expressions.StringLocate",
+      // _FUNC_ is replaced by `%` which causes a parsing error on `SELECT %(2, 1.8)`
+      "org.apache.spark.sql.catalyst.expressions.Remainder",
+      // Examples demonstrate alternative names, see SPARK-20749
+      "org.apache.spark.sql.catalyst.expressions.Length")
+    spark.sessionState.functionRegistry.listFunction().foreach { funcId =>
+      val info = spark.sessionState.catalog.lookupFunctionInfo(funcId)
+      val className = info.getClassName
+      withClue(s"Expression class '$className'") {
+        val exprExamples = info.getOriginalExamples
+        if (!exprExamples.isEmpty && !ignoreSet.contains(className)) {
+          assert(exampleRe.findAllIn(exprExamples).toSet.forall(_.contains("_FUNC_")))
+        }
+      }
+    }
+  }
+
   test("SPARK-6743: no columns from cache") {
     Seq(
       (83, 0, 38),