Decimal cast to smaller precision gives invalid (off-by-one) result in some cases

[Blizzara]

Describe the bug

See practical examples in apache/datafusion#14450.
Somehow a regression caused by #6833.

To Reproduce

Adding this into arrow-cast/src/cast/mod.rs:

    #[test]
    fn test_decimal_to_decimal() {
        let array = vec![Some(520)];
        let array = create_decimal_array(array, 4, 2).unwrap();
        let input_type = DataType::Decimal128(4, 2);
        let output_type = DataType::Decimal128(3, 2);
        assert!(can_cast_types(&input_type, &output_type));

        let options = CastOptions {
            safe: false,
            ..Default::default()
        };
        let result = cast_with_options(&array, &output_type, &options);
        assert_eq!(result.unwrap().as_primitive::<Decimal128Type>().value(0), 520);
    }

passes on 84dba34, but fails on eb7ab83 with

assertion `left == right` failed
  left: 521
 right: 520

Somehow something in the code adds an extra 1 to the value?

Expected behavior

Test passes

Additional context

Fyi @himadripal @andygrove @findepi @viirya @tustvold as you were involved in that earlier change. I tried to look at it but don't understand how it'd go wrong myself, maybe you know?

[himadripal]

@Blizzara I'll take a look, as part of the previous PR, I added validation after creation like it was for float, wondering if validation message would add that extra 1?

[Blizzara]

Actually @himadripal I think the issue is this line: eb7ab83#diff-07a573af92ff5330f77407c559e89e06b02ef8c466a394ca86a710023f000593R170, the < should be probably <= to match the convert_to_bigger_or_equal_scale_decimal. If I do that change then my test passes. Does that make sense to you?

[himadripal]

@Blizzara it make sense, when input scale = output scale it always goes to convert_to_smaller_scale_decimal instead it should go to convert_to_bigger_or_equal_scale_decimal , I think, tests added as part of the #6833 needs change . Let me know if you are going to make this change, I'll create a PR otherwise to fix this now.

[himadripal]

@Blizzara Please check.