Improve variable naming in writeResults function

[rhnvrm]

Summary

Improve variable naming and documentation in the writeResults function to make the SQL row scanning logic clearer and more maintainable.

Changes

File: internal/core/core.go

Variable Renaming:

• resCols → rowValues (holds actual data values)
• resPointers → scanPointers (holds pointers for scanning)

Documentation Update:

• Replaced unclear comment with detailed explanation of the indirection pattern
• Added clear documentation explaining why the pointer pattern is needed

Before:

// Gymnastics to read arbitrary types from the row.
var (
    resCols     = make([]interface{}, numCols)
    resPointers = make([]interface{}, numCols)
)

After:

// Create indirection for scanning arbitrary column types.
// sql.Rows.Scan() requires pointers to write into, but we need the actual values for WriteRow().
// Solution: scanPointers holds pointers to each element in rowValues, allowing Scan() to
// write through the pointers while we use the actual values from rowValues.
var (
    rowValues    = make([]interface{}, numCols)
    scanPointers = make([]interface{}, numCols)
)

Impact

• ✅ No functional changes - pure readability improvement
• ✅ Clearer variable names that express their purpose
• ✅ Better documentation explaining the data flow pattern
• ✅ Easier code maintenance and understanding

The variable names now clearly express that one slice holds the actual data while the other holds pointers for the scanning operation.