Fix swapped arithmetic crystal class numbers for space groups 187-190

[lan496]

Summary

Moyo has the same data bug as spglib, fixed there in spglib/spglib#632: the arithmetic crystal class (ACC) numbers 56 and 57 are swapped relative to International Tables for Crystallography, Volume C (2004), Table 1.4.2.1.

G	Correct ACC	Moyo before this PR
187: P-6m2	56: -6m2P	57: -6m2P
188: P-6c2	56: -6m2P	57: -6m2P
189: P-62m	57: -62mP	56: -62mP
190: P-62c	57: -62mP	56: -62mP

The swap was applied consistently across the internal tables, so symmetry identification and the reported ACC symbols were already correct; only the ACC numbers exposed via MoyoDataset / moyopy.SpaceGroupType.arithmetic_number were wrong.

Changes:

• moyo/src/data/arithmetic_crystal_class.rs: swap the symbols so 56 = -6m2P and 57 = -62mP
• moyo/src/data/hall_symbol_database.rs: Hall entries 481/482 (SG 187/188) now map to ACC 56, and 483/484 (SG 189/190) to ACC 57
• moyo/src/data/point_group.rs: PointGroupRepresentative::from_arithmetic_crystal_class maps 56 -> Hall 481 and 57 -> Hall 483
• CHANGELOG.md: Unreleased entry

Test plan

• New regression test test_arithmetic_crystal_class_for_sg_187_to_190 asserting the ACC number and symbol for every Hall entry of SG 187-190 against Table 1.4.2.1 (fails before the fix, passes after)
• cargo test -p moyo passes (161 unit + integration tests)
• just prek passes

🤖 Generated with Claude Code