Support for calling stored prcoedure with array parameter

[Earammak]

Implemented a new API to fetch the results of a stored procedure call using the prepare → bind → execute flow, with support for both scalar and array parameters.

Scalar support:
For procedures with single‑value arguments, ibm_db.fetch_callproc works seamlessly with ibm_db.callproc.
Scalars are bound using ibm_db.bind_param() with SQL_PARAM_INPUT, SQL_PARAM_OUTPUT, or SQL_PARAM_INOUT.
Python values are mapped directly to SQL types.

Array support:
For procedures with array arguments, the implementation applies correct cardinality and buffer allocation using SQLSetDescField() with SQL_DESC_CARDINALITY and SQL_DESC_CARDINALITY_PTR.

INPUT arguments use the Implementation Parameter Descriptor (hIPD).

OUTPUT arguments use the Application Parameter Descriptor (hAPD).

INOUT arguments configure both hIPD and hAPD.

This logic is executed internally after SQLBindParameter, ensuring that both scalar and array parameters are properly bound before execution.

After execution, ibm_db.fetch_callproc retrieves updated parameter values.
For input parameters return the same values passed in and for OUT/INOUT parameters return the modified values produced by the stored procedure.

This enhancement allows to pass either Python lists (for arrays) or scalars (for single values) as parameters and receive the updated values after execution as a tuple containing the statement object followed by the parameters passed to the procedure.