Enable customizable precision for allocations

[faraquet]

Problem

Me and @spariev have found that the allocation logic fails to function properly with large arrays. In a real production scenario, we encountered a process freeze followed by an error.

Money::Allocation.generate(246.4, [
  81.29, 81.29, 81.29, 81.29,
  234.8, 234.8, 234.8, 234.8,
  90.36, 90.36, 90.36, 90.36,
  90.36, 90.36, 90.36, 90.36,
  90.36, 90.36, 90.36, 90.36,
  90.36, 90.36, 90.36, 90.36,
  90.36, 90.36, 90.36, 90.36,
  90.36
], false)

The reason is that when the number of parts exceeds 15, there is an infinite number of digits after the decimal point.

Solution

The generate method has been refactored by renaming the whole_amounts parameter to rounding_mode for better clarity. This allows more flexible rounding options, supporting booleans and custom decimal precision.

[faraquet]

@semmons99 @antstorm
Hi 👋 , could you please take a look? I'm not sure who is more responsible for reviewing, so I'm tagging you because you approved the recent commits. I'm happy to discuss these changes. Thanks!