Handle overflow precision when casting from integer to decimal

[viirya]

Describe the bug

Although cast kernel uses CastOptions when casting from integer to decimal, its behavior on overflow-checking is incomplete.

For now, it'd check division or multiplication if overflowing on underlying decimal native types (i.e., i128 and i256). But it doesn't check if input integer is overflowing on decimal precision. This causes weird behavior when downstream project (e.g., DataFusion) uses cast kernel to cast integer to decimal.

To Reproduce

Expected behavior

Additional context

[tustvold]

label_issue.py automatically added labels {'arrow'} from #3996