fix: Implement Substrait consumer support for like_match, like_imatch, and negated variants (#18929)

## Which issue does this PR close?

<!--
We generally require a GitHub issue to be filed for all bug fixes and
enhancements and this helps us generate change logs for our releases.
You can link an issue to this PR using the GitHub syntax. For example
`Closes #123` indicates that this PR will close issue #123.
-->

- Closes #16293 

## Rationale for this change

<!--
Why are you proposing this change? If this is already explained clearly
in the issue then this section is not needed.
Explaining clearly why changes are proposed helps reviewers understand
your changes and offer better suggestions for fixes.
-->

Currently, DataFusion fails to consume Substrait plans that utilize
`like_match`, `like_imatch`, or their negated variants
(`like_not_match`, `like_not_imatch`). This results in a panic with
`DataFusion error: This feature is not implemented: Unsupported function
name: "like_match"` during round-trip planning.

This PR implements the missing mapping logic in the Substrait consumer
to correctly translate these function names into DataFusion `Expr::Like`
expressions.

## What changes are included in this PR?

<!--
There is no need to duplicate the description in the issue here but it
is sometimes worth providing a summary of the individual changes in this
PR.
-->

- Updated `BuiltinExprBuilder` in `scalar_function.rs` to recognize:
  - `like_match` (LIKE)
  - `like_imatch` (ILIKE)
  - `like_not_match` (NOT LIKE)
  - `like_not_imatch` (NOT ILIKE)
- Modified the internal `build_like_expr` helper to accept a `negated`
boolean flag (previously hardcoded to `false`).
- Added a new unit test `test_like_match_conversion` to verify the
correct conversion of these functions and their arguments.

## Are these changes tested?

<!--
We typically require tests for all PRs in order to:
1. Prevent the code from being accidentally broken by subsequent changes
2. Serve as another way to document the expected behavior of the code

If tests are not included in your PR, please explain why (for example,
are they covered by existing tests)?
-->

Yes. I added a new unit test `test_like_match_conversion` in
`datafusion/substrait/src/logical_plan/consumer/expr/scalar_function.rs`.

This test explicitly constructs Substrait plans for `like_match` and
`like_not_match` and asserts that they are consumed into correct
DataFusion `Expr::Like` variants with the appropriate `negated` and
`case_insensitive` flags.

**Note: Verification was done via unit test rather than the standard
`sqllogictest` suite because `predicates.slt` was experiencing unrelated
environment-specific failures locally.**

## Are there any user-facing changes?

<!--
If there are user-facing changes then we may require documentation to be
updated before approving the PR.
-->

<!--
If there are any breaking changes to public APIs, please add the `api
change` label.
-->

- No breaking changes.
- Users relying on Substrait round-trips for LIKE/ILIKE expressions will
now see these queries succeed instead of failing.

Author: Nithurshen

datafusion/substrait/src/logical_plan/consumer/expr/scalar_function.rs

@@ -177,8 +177,9 @@ struct BuiltinExprBuilder {
 impl BuiltinExprBuilder {
     pub fn try_from_name(name: &str) -> Option<Self> {
         match name {
-            "not" | "like" | "ilike" | "is_null" | "is_not_null" | "is_true"
-            | "is_false" | "is_not_true" | "is_not_false" | "is_unknown"
+            "not" | "like" | "ilike" | "like_match" | "like_imatch"
+            | "like_not_match" | "like_not_imatch" | "is_null" | "is_not_null"
+            | "is_true" | "is_false" | "is_not_true" | "is_not_false" | "is_unknown"
             | "is_not_unknown" | "negative" | "negate" | "and_not" | "xor"
             | "between" | "logb" => Some(Self {
                 expr_name: name.to_string(),
@@ -194,8 +195,12 @@ impl BuiltinExprBuilder {
         args: Vec<Expr>,
     ) -> Result<Expr> {
         match self.expr_name.as_str() {
-            "like" => Self::build_like_expr(false, f, args).await,
-            "ilike" => Self::build_like_expr(true, f, args).await,
+            "like" => Self::build_like_expr(false, false, f, args).await,
+            "ilike" => Self::build_like_expr(true, false, f, args).await,
+            "like_match" => Self::build_like_expr(false, false, f, args).await,
+            "like_imatch" => Self::build_like_expr(true, false, f, args).await,
+            "like_not_match" => Self::build_like_expr(false, true, f, args).await,
+            "like_not_imatch" => Self::build_like_expr(true, true, f, args).await,
             "not" | "negative" | "negate" | "is_null" | "is_not_null" | "is_true"
             | "is_false" | "is_not_true" | "is_not_false" | "is_unknown"
             | "is_not_unknown" => Self::build_unary_expr(&self.expr_name, args).await,
@@ -236,6 +241,7 @@ impl BuiltinExprBuilder {
 
     async fn build_like_expr(
         case_insensitive: bool,
+        negated: bool,
         f: &ScalarFunction,
         args: Vec<Expr>,
     ) -> Result<Expr> {
@@ -274,7 +280,7 @@ impl BuiltinExprBuilder {
         };
 
         Ok(Expr::Like(Like {
-            negated: false,
+            negated,
             expr: Box::new(expr),
             pattern: Box::new(pattern),
             escape_char,
@@ -476,4 +482,96 @@ mod tests {
         let _ = consumer.consume_scalar_function(&func, &df_schema).await?;
         Ok(())
     }
+
+    #[tokio::test]
+    async fn test_like_match_conversion() -> Result<()> {
+        // 1. Setup the consumer with the "like_match" function registered
+        let mut extensions = Extensions::default();
+        extensions
+            .functions
+            .insert(0, "like_match:str_str".to_string());
+        extensions
+            .functions
+            .insert(1, "like_not_match:str_str".to_string());
+        extensions
+            .functions
+            .insert(2, "like_imatch:str_str".to_string());
+
+        let consumer = DefaultSubstraitConsumer::new(&extensions, &TEST_SESSION_STATE);
+
+        // 2. Create the arguments (column "a" and pattern "%foo%")
+        let schema = Schema::new(vec![Field::new("a", DataType::Utf8, false)]);
+        let df_schema = DFSchema::try_from(schema).unwrap();
+
+        let col_arg = FunctionArgument {
+            arg_type: Some(ArgType::Value(Expression {
+                rex_type: Some(RexType::Selection(Box::new(
+                    substrait::proto::expression::FieldReference {
+                        reference_type: Some(substrait::proto::expression::field_reference::ReferenceType::DirectReference(
+                            substrait::proto::expression::ReferenceSegment {
+                                reference_type: Some(substrait::proto::expression::reference_segment::ReferenceType::StructField(
+                                    Box::new(substrait::proto::expression::reference_segment::StructField {
+                                        field: 0,
+                                        child: None,
+                                    })
+                                )),
+                            }
+                        )),
+                        root_type: Some(substrait::proto::expression::field_reference::RootType::RootReference(
+                            substrait::proto::expression::field_reference::RootReference {}
+                        )),
+                    }
+                ))),
+            })),
+        };
+
+        let pattern_arg = FunctionArgument {
+            arg_type: Some(ArgType::Value(Expression {
+                rex_type: Some(RexType::Literal(Literal {
+                    nullable: false,
+                    type_variation_reference: 0,
+                    literal_type: Some(LiteralType::String("foo".to_string())),
+                })),
+            })),
+        };
+
+        // 3. Test "like_match" (Standard LIKE)
+        let func_like = ScalarFunction {
+            function_reference: 0,
+            arguments: vec![col_arg.clone(), pattern_arg.clone()],
+            ..Default::default()
+        };
+
+        let result = consumer
+            .consume_scalar_function(&func_like, &df_schema)
+            .await?;
+
+        if let Expr::Like(like) = result {
+            assert!(!like.negated);
+            assert!(!like.case_insensitive);
+            assert_eq!(format!("{}", like.pattern), "Utf8(\"foo\")");
+        } else {
+            panic!("Expected Expr::Like, got {result:?}");
+        }
+
+        // 4. Test "like_not_match" (NOT LIKE)
+        let func_not_like = ScalarFunction {
+            function_reference: 1,
+            arguments: vec![col_arg.clone(), pattern_arg.clone()],
+            ..Default::default()
+        };
+
+        let result = consumer
+            .consume_scalar_function(&func_not_like, &df_schema)
+            .await?;
+
+        if let Expr::Like(like) = result {
+            assert!(like.negated);
+            assert!(!like.case_insensitive);
+        } else {
+            panic!("Expected Expr::Like (negated), got {result:?}");
+        }
+
+        Ok(())
+    }
 }