ROW data type for stored routine variables

Add support for the ROW data type variables in stored routines according to this SQL Standard syntax:

<row type> ::= ROW <row type body>

<row type body> ::= <left paren> <field definition> [ { <comma> <field definition> }... ] <right paren>

<field definition> ::= <field name> <data type>

<data type> ::= <predefined type>

and add support for a new expression type, a ROW field reference, as follows:

<field reference> ::= <row variable> <period> <field name>

where <row variable> is a stored routine variable declared using <row type>.

Example:

DELIMITER $$

CREATE PROCEDURE p1()

BEGIN

  DECLARE a ROW (c1 INT, c2 VARCHAR(10));

  SET a.c1= 10;

  SET a.c2= 'test';

  INSERT INTO t1 VALUES (a.c1, a.c2);

END;

$$

DELIMITER ;

CALL p1();

This task is needed to create infrastructure for MDEV-10593

This task is also needed as a prerequisite for MDEV-10581, where this statement:

FOR rec IN cursor

will automatically declare an index variable rec of the ROW data type, according to the cursor structure.

Row data type features:

• Declaration of a ROW type stored routine variable (both local variables and parameters)
• Declaration of a ROW type stored procedure OUT parameter
• Default values in a ROW type variable declaration (e.g. DEFAULT ROW(1,2))
• Assignment of a ROW type variable from another ROW type variable (using the SET command and the := operator in sql_mode=ORACLE)
• Assignment of a ROW type variable from a ROW() function result (using the SET command and the := operator in sql_mode=ORACLE)
• Passing a ROW type variable and a ROW() function result to stored routines
• Comparison of a ROW type variable to another ROW type variable
• Comparison of a ROW type variable to ROW() function

Row field features:

ROW fields (members) will act as normal variables, and will be able to appear in all query parts where an SP variable is allowed:

• assignment (using the SET command)

  SET a.x= 10, a.y=20, a.z= b.z;

• assignment (sql_mode=ORACLE specific syntax)

  a.x:= 10;

  a.x:= b.x;

• passing to functions and operators

  SELECT f1(rec.a), rec.a<10;

• clauses: select list, WHERE, HAVING, LIMIT, etc

  SELECT var.a, t1.b FROM t1 WHERE t1.b=var.b LIMIT var.c;

• INSERT values

  INSERT INTO t1 VALUES (rec.a, rec.b, rec.c);

• SELECT .. INTO targets

  SELECT a,b INTO rec.a, rec.b FROM t1 WHERE t1.id=10;

• Dynamic SQL out parameters (EXECUTE and EXECUTE IMMEDIATE)

  EXECUTE IMMEDIATE 'CALL proc_with_out_param(?)' USING rec.a;

Features not supported:

The following features are out of scope of this task and will be implemented separately:

• Returning a ROW type expression from a stored function (see MDEV-12252). This will need some grammar change to support field names after parentheses:

  SELECT f1().x FROM DUAL;

• Returning a ROW type expression from a built-in hybrid type function, such as CASE, IF, etc.
• ROW of ROWs