chore[dtype]: use NonZero for decimal precision #5146

Performance Gate Passed

⚡ 10 improvements
✅ 1308 untouched
🆕 7 new
⏩ 120 skipped¹

	Benchmark	`BASE`	`HEAD`	Change
⚡	`new_alp_prim_test_between[f32, 16384]`	104.8 µs	90.5 µs	+15.9%
⚡	`new_alp_prim_test_between[f64, 16384]`	137.7 µs	107.4 µs	+28.14%
⚡	`old_alp_prim_test_between[f32, 16384]`	175.2 µs	158.9 µs	+10.27%
⚡	`old_alp_prim_test_between[f32, 32768]`	339.2 µs	292.6 µs	+15.91%
⚡	`old_alp_prim_test_between[f64, 16384]`	215 µs	181.3 µs	+18.57%
⚡	`old_alp_prim_test_between[f64, 32768]`	328.1 µs	262.6 µs	+24.92%
⚡	`new_bp_prim_test_between[i64, 32768]`	205 µs	158.2 µs	+29.61%
⚡	`old_bp_prim_test_between[i32, 32768]`	229.4 µs	200.4 µs	+14.43%
⚡	`old_bp_prim_test_between[i64, 16384]`	195.1 µs	166.1 µs	+17.51%
⚡	`old_bp_prim_test_between[i64, 32768]`	303.8 µs	241.7 µs	+25.72%
🆕	`decompress[("alp_for_bp_f64", 0x765300)]`	N/A	24.2 ms	N/A
🆕	`decompress[("datetime_for_bp", 0x7685e0)]`	N/A	34.9 ms	N/A
🆕	`decompress[("dict_fsst_varbin_bp_string", 0x7676f0)]`	N/A	14.5 ms	N/A
🆕	`decompress[("dict_fsst_varbin_string", 0x767060)]`	N/A	14.5 ms	N/A
🆕	`decompress[("dict_varbinview_string", 0x765d20)]`	N/A	14.7 ms	N/A
🆕	`decompress[("for_bp_u64", 0x764bb0)]`	N/A	2.5 ms	N/A
🆕	`decompress[("runend_for_bp_u32", 0x7661b0)]`	N/A	2 ms	N/A

120 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩