parser: support SUBPARTITION grammar for table partition

[tiancaiamao]

What problem does this PR solve?

There are many test cases in mysql-test use subpartition.
If we parse the subpartition grammar, it would be more convenient for us to use the test cases.

CREATE TABLE t1 (
    ID int(11) NOT NULL,
    `aaaa,aaaaa` tinyint(3) UNSIGNED NOT NULL DEFAULT '0',
    ddddddddd int(11) NOT NULL DEFAULT '0',
    new_field0 varchar(50),
    PRIMARY KEY(ID, `aaaa,aaaaa`, ddddddddd))
PARTITION BY RANGE(ID)
PARTITIONS 3
SUBPARTITION BY LINEAR KEY(ID,`aaaa,aaaaa`)
SUBPARTITIONS 2 (
    PARTITION p01 VALUES LESS THAN(100),
    PARTITION p11 VALUES LESS THAN(200),
    PARTITION p21 VALUES LESS THAN MAXVALUE);

What is changed and how it works?

Parser will parse create table ... partition by range ... subpartition ...

Check List

Tests

• Unit test

Code changes

• update parser.y add the rules
• add "subpartition" and "subpartitions" to lexer
• add test

@coocood @zhexuany @winkyao

[jackysp]

LGTM

[imtbkcat]

LGTM

[imtbkcat]

/run-all-tests

[XuHuaiyu]

LGTM