YJIT: A64: Avoid intermediate register in opt_and and friends

[XrXr]

Same idea as the x64 equivalent in c2622b5, removing the register
shuffle coming from the pop two, push one stack motion these VM
instructions perform.

# Insn: 0004 opt_or (stack_size: 2)
- orr x11, x1, x9
- mov x1, x11
+ orr x1, x1, x9

[launchable-app]

Launchable Report

❌ Test session #2806658 failed   details on CI
_{🔔 no issues ✖️1 test failed ✔️25737 tests passed}

Passed test sessions

✅ Test session #2806642 passed   details on CI
✅ Test session #2806643 passed   details on CI
✅ Test session #2806644 passed   details on CI
✅ Test session #2806646 passed   details on CI
✅ Test session #2806647 passed   details on CI
✅ Test session #2806649 passed   details on CI
✅ Test session #2806651 passed   details on CI
✅ Test session #2806653 passed   details on CI
✅ Test session #2806654 passed   details on CI
✅ Test session #2806656 passed   details on CI
✅ Test session #2806665 passed   details on CI
✅ Test session #2806676 passed   details on CI
✅ Test session #2806677 passed   details on CI
✅ Test session #2806828 passed   details on CI

Build: refs_pull_10509_merge_e29e6673ff7cb190f1f44491ed3792714b995bd2