PR Summary

• Improved Data Type Conversion Functionality
  We have improved the 'check_cast' function, which is used to convert or translate data from one type to another. This enhancement now allows for more diverse and extra conversion rules among various data types.

• Incorporated Detailed Scenario Handling
  A comprehensive set of rules (referred to as a match statement) has been designed and implemented to deal with specialized conversion scenarios. This includes the different cases such as 'Null', 'Nullable', 'Array', and 'Map' data types.

• Enhanced Conversion between 'Variant' and 'Map', and between 'Tuple' types
  The functionality to convert from 'Variant' to 'Map' (with tuple type) and from one 'Tuple' to another 'Tuple' has been added. The only condition for this conversion is that the source and destination tuples should have an equal number of data type components.

• Robust Error Handling
  In the event of a situation where data conversion is not feasible or supported, our improved error handling mechanisms will produce a detailed, meaningful error message. This feature helps in understanding what went wrong and aids in troubleshooting.

please check the ci

please add logic tests for the pr.

create table a(t decimal(15,2));
 create table b (t json);


 insert into a select * from b;  --- should fail during binding 

seems like insert into a select * from b; will not trigger type_check()...

seems like insert into a select * from b; will not trigger type_check()...

TransformCastSchema will trigger

let's use explain pipeline ...