Show join predicates in vt keys and vt benchstat

[frouioui]

Descrition

This PR is based on #42. And the vitess version used in the PR is based on vitessio/vitess#17130.

With this PR we can now list and summarize the usage of Join Predicate per table and per query. This will allow us to know exactly what join predicates we use per table/query and understand the relationship between the different tables present in the query log.

Results

For the following query:

select c.customer_id, p.area_name from customers c join pincode_areas p on c.customer_pincode = p.pincode and c.customer_pincode = p.area_name;

We get the following output with vt keys:

      {
        "queryStructure": "SELECT `c`.`customer_id`, `p`.`area_name` FROM `customers` AS `c` JOIN `pincode_areas` AS `p` ON `c`.`customer_pincode` = `p`.`pincode` AND `c`.`customer_pincode` = `p`.`area_name`",
        "usageCount": 1,
        "lineNumbers": [
          131
        ],
        "tableName": [
          "customers",
          "pincode_areas"
        ],
        "joinColumns": [
          "customers.customer_pincode =",
          "pincode_areas.area_name =",
          "pincode_areas.pincode ="
        ],
        "joinPredicates": [
          "customers.customer_pincode = pincode_areas.pincode",
          "customers.customer_pincode = pincode_areas.area_name"
        ],
        "statementType": "SELECT"
      },

Which can then be fed to vt benchstat, and we get:

Table: customers used in 2 queries
+------------------+----------+------------+--------+
|      Column      | Filter % | Grouping % | Join % |
+------------------+----------+------------+--------+
| customer_pincode | 0.00%    | 0.00%      | 50.00% |
+------------------+----------+------------+--------+
+------------------------------------------------------+
|                    Join Predicate                    |
+------------------------------------------------------+
| customers.customer_pincode = pincode_areas.pincode   |
| customers.customer_pincode = pincode_areas.area_name |
+------------------------------------------------------+

Table: orders used in 1 queries
+--------+----------+------------+--------+
| Column | Filter % | Grouping % | Join % |
+--------+----------+------------+--------+
+--------+----------+------------+--------+
+----------------+
| Join Predicate |
+----------------+
+----------------+

Table: pincode_areas used in 2 queries
+-----------+----------+------------+--------+
|  Column   | Filter % | Grouping % | Join % |
+-----------+----------+------------+--------+
| area_name | 0.00%    | 0.00%      | 50.00% |
| pincode   | 0.00%    | 0.00%      | 50.00% |
+-----------+----------+------------+--------+
+------------------------------------------------------+
|                    Join Predicate                    |
+------------------------------------------------------+
| customers.customer_pincode = pincode_areas.pincode   |
| customers.customer_pincode = pincode_areas.area_name |
+------------------------------------------------------+