Fix stuttering GUC in error/warning message (#844)

Running the command in the error/warning message would fail or do
nothing. This makes it use the right command.

Author: JelteF

src/pgduckdb_types.cpp

@@ -1289,20 +1289,20 @@ ConvertPostgresToBaseDuckColumnType(Form_pg_attribute &attribute) {
 			if (type_modifier == -1) {
 				return duckdb::LogicalType::USER(
 				    "DuckDB requires the precision of a NUMERIC to be set. You can choose to convert these NUMERICs to "
-				    "a DOUBLE by using 'SET duckdb.duckdb.convert_unsupported_numeric_to_double = true'");
+				    "a DOUBLE by using 'SET duckdb.convert_unsupported_numeric_to_double = true'");
 			} else if (precision < 1 || precision > 38) {
 				return duckdb::LogicalType::USER(
 				    "DuckDB only supports NUMERIC with a precision of 1-38. You can choose to convert these NUMERICs "
-				    "to a DOUBLE by using 'SET duckdb.duckdb.convert_unsupported_numeric_to_double = true'");
+				    "to a DOUBLE by using 'SET duckdb.convert_unsupported_numeric_to_double = true'");
 			} else if (scale < 0 || scale > 38) {
 				return duckdb::LogicalType::USER(
 				    "DuckDB only supports NUMERIC with a scale of 0-38. You can choose to convert these NUMERICs to a "
-				    "DOUBLE by using 'SET duckdb.duckdb.convert_unsupported_numeric_to_double = true'");
+				    "DOUBLE by using 'SET duckdb.convert_unsupported_numeric_to_double = true'");
 			} else {
 				return duckdb::LogicalType::USER(
 				    "DuckDB does not support NUMERIC with a scale that is larger than the precision. You can choose to "
-				    "convert these NUMERICs to a DOUBLE by using 'SET "
-				    "duckdb.duckdb.convert_unsupported_numeric_to_double = true'");
+				    "convert these NUMERICs to a DOUBLE by using 'SET duckdb.convert_unsupported_numeric_to_double = "
+				    "true'");
 			}
 		}
 

test/regression/expected/array_type_support.out

@@ -308,7 +308,7 @@ INSERT INTO numeric_array_1d SELECT CAST(a as NUMERIC[]) FROM (VALUES
     ('{}')
 ) t(a);
 SELECT * FROM numeric_array_1d;
-WARNING:  Unsupported Postgres type: DuckDB requires the precision of a NUMERIC to be set. You can choose to convert these NUMERICs to a DOUBLE by using 'SET duckdb.duckdb.convert_unsupported_numeric_to_double = true'
+WARNING:  Unsupported Postgres type: DuckDB requires the precision of a NUMERIC to be set. You can choose to convert these NUMERICs to a DOUBLE by using 'SET duckdb.convert_unsupported_numeric_to_double = true'
          a          
 --------------------
  {1.1,2.2,3.3}
@@ -596,7 +596,7 @@ INSERT INTO numeric_array_2d VALUES
     ('{}'),
     ('{{11.1,12.2},{NULL,14.4}}');
 SELECT * FROM numeric_array_2d;
-WARNING:  Unsupported Postgres type: DuckDB requires the precision of a NUMERIC to be set. You can choose to convert these NUMERICs to a DOUBLE by using 'SET duckdb.duckdb.convert_unsupported_numeric_to_double = true'
+WARNING:  Unsupported Postgres type: DuckDB requires the precision of a NUMERIC to be set. You can choose to convert these NUMERICs to a DOUBLE by using 'SET duckdb.convert_unsupported_numeric_to_double = true'
                a                
 --------------------------------
  {{1.1,2.2},{3.3,4.4}}

test/regression/expected/temporary_tables.out

@@ -165,7 +165,7 @@ CREATE TEMP TABLE t(a int) WITH (fillfactor = 50);
 ERROR:  Storage options are not supported in DuckDB
 -- Should fail because user should specify the precision of the NUMERIC.
 CREATE TEMP TABLE large_numeric_tbl (a NUMERIC) USING duckdb;
-ERROR:  (PGDuckDB/duckdb_create_table_trigger_cpp) Not implemented Error: Unsupported Postgres type: DuckDB requires the precision of a NUMERIC to be set. You can choose to convert these NUMERICs to a DOUBLE by using 'SET duckdb.duckdb.convert_unsupported_numeric_to_double = true'
+ERROR:  (PGDuckDB/duckdb_create_table_trigger_cpp) Not implemented Error: Unsupported Postgres type: DuckDB requires the precision of a NUMERIC to be set. You can choose to convert these NUMERICs to a DOUBLE by using 'SET duckdb.convert_unsupported_numeric_to_double = true'
 -- But it's fine if the user specifies the precision
 CREATE TEMP TABLE large_numeric_tbl_specified (a NUMERIC(38,20)) USING duckdb;
 -- CTAS is fine though, it will use duckdb its default

test/regression/expected/type_support.out

@@ -327,7 +327,7 @@ INSERT INTO numeric_as_double SELECT a FROM (VALUES
     (458234502034234234234.000012)
 ) t(a);
 SELECT * FROM numeric_as_double;
-WARNING:  Unsupported Postgres type: DuckDB requires the precision of a NUMERIC to be set. You can choose to convert these NUMERICs to a DOUBLE by using 'SET duckdb.duckdb.convert_unsupported_numeric_to_double = true'
+WARNING:  Unsupported Postgres type: DuckDB requires the precision of a NUMERIC to be set. You can choose to convert these NUMERICs to a DOUBLE by using 'SET duckdb.convert_unsupported_numeric_to_double = true'
               a               
 ------------------------------
                   0.234234234