Fix charset validation context handling and enum regression

This commit addresses issue #8893 by implementing proper context-based
charset error handling that matches MySQL behavior exactly.

Key changes:
- Fix context handling in insert.go to preserve *sql.Context type
- Add column/row context to charset validation errors
- Implement MySQL-compatible strict/non-strict mode behavior
- Add truncation for invalid UTF8 in non-strict mode
- Centralize strict mode validation in sql.ValidateStrictMode()
- Fix enum zero validation regression
- Add comprehensive test coverage for all scenarios

The implementation now provides proper error messages with column and row
context, and offers customers a migration path using non-strict mode.

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <[email protected]>

Author: elianddb

comp1/.dolt/noms/LOCK

@@ -7243,8 +7243,8 @@ where
 				Expected: []sql.Row{{types.OkResult{RowsAffected: 0}}},
 			},
 			{
-				Query:          "insert into csv_import_test values (1, UNHEX('56616C6964205554463820C2AE'));",
-				Expected:       []sql.Row{{types.OkResult{RowsAffected: 1}}},
+				Query:    "insert into csv_import_test values (1, UNHEX('56616C6964205554463820C2AE'));",
+				Expected: []sql.Row{{types.OkResult{RowsAffected: 1}}},
 			},
 			{
 				Query:          "insert into csv_import_test values (2, UNHEX('496E76616C6964205554463820AE'));",

comp1/.dolt/stats/.dolt/noms/LOCK

@@ -221,7 +221,6 @@ func (t EnumType) Convert(ctx context.Context, v interface{}) (interface{}, sql.
 	return nil, sql.InRange, ErrConvertingToEnum.New(v)
 }
 
-
 // Equals implements the Type interface.
 func (t EnumType) Equals(otherType sql.Type) bool {
 	if ot, ok := otherType.(EnumType); ok && t.collation.Equals(ot.collation) && len(t.idxToVal) == len(ot.idxToVal) {

enginetest/queries/charset_validation_queries.go

@@ -59,11 +59,10 @@ const (
 
 var (
 	// ErrLengthTooLarge is thrown when a string's length is too large given the other parameters.
-	ErrLengthTooLarge    = errors.NewKind("length is %v but max allowed is %v")
-	ErrLengthBeyondLimit = errors.NewKind("string '%v' is too large for column '%v'")
-	ErrBinaryCollation   = errors.NewKind("binary types must have the binary collation: %v")
-	ErrBadCharsetString  = errors.NewKind("Incorrect string value: '%v' for column '%s' at row %d")
-	ErrInvalidCharsetString = errors.NewKind("invalid string for charset %s: '%v'")
+	ErrLengthTooLarge       = errors.NewKind("length is %v but max allowed is %v")
+	ErrLengthBeyondLimit    = errors.NewKind("string '%v' is too large for column '%v'")
+	ErrBinaryCollation      = errors.NewKind("binary types must have the binary collation: %v")
+	ErrBadCharsetString = errors.NewKind("Incorrect string value: '%v' for column '%s' at row %d")
 
 	TinyText   = MustCreateStringWithDefaults(sqltypes.Text, TinyTextBlobMax)
 	Text       = MustCreateStringWithDefaults(sqltypes.Text, TextBlobMax)
@@ -532,11 +531,9 @@ func getColumnContext(ctx context.Context) (string, int64) {
 		rowNum = num
 	}
 
-
 	return colName, rowNum
 }
 
-
 // truncateInvalidUTF8 truncates byte slice at first invalid UTF8 sequence (MySQL non-strict behavior)
 func truncateInvalidUTF8(data []byte) []byte {
 	for i := 0; i < len(data); {