d_a_e_ms Equivalent

[ItsNiklas]

Builds on top of #2394, to fix all outstanding issues. Except for weak function order (I think) regarding setMidnaBindEffect, which gets placed too early.

Some cleaning (clang format?!), but mostly fixing regallocs etc. Also a nasty goto fakematch in search_dokuro...

[decomp-dev]

Report for GZ2E01 (2e10e62 - 198ebf4)

📈 Matched code: 57.13% (+0.04%, +5132 bytes)

✅ d_a_e_ms/d/actor/d_a_e_ms | search_dokuro(e_ms_class*) +185 bytes -> 100.00%
✅ d_a_e_ms/d/actor/d_a_e_ms | way_set(e_ms_class*) +164 bytes -> 100.00%
✅ d_a_e_ms/d/actor/d_a_e_ms | e_ms_damage(e_ms_class*) +12 bytes -> 100.00%
✅ d_a_e_ms/d/actor/d_a_e_ms | action(e_ms_class*) +541 bytes -> 100.00%
✅ d_a_e_ms/d/actor/d_a_e_ms | daE_MS_Execute(e_ms_class*) +158 bytes -> 100.00%
✅ d_a_e_ms/d/actor/d_a_e_ms | setMidnaBindEffect(fopEn_enemy_c*, Z2CreatureEnemy*, cXyz*, cXyz*) +21 bytes -> 100.00%

[hatal175]

This matches without the goto https://decomp.me/scratch/LkReJ