Legacy unprefixed voice-mode rows were dropped during the #12542
platform-namespacing migration, causing one unwanted auto-TTS reply on
Telegram chats that had /voice off before the upgrade. Preserve legacy
"off" rows on load and apply them as a cross-platform fallback; explicit
platform-prefixed rows take precedence so a later /voice on/tts cannot
be silently re-suppressed after a restart.
Fixes#14025
Follow-up to 40164ba1.
- _handle_voice_channel_join/leave now use event.source.platform instead of
hardcoded Platform.DISCORD (consistent with other voice handlers).
- Update tests/gateway/test_voice_command.py to use 'platform:chat_id' keys
matching the new _voice_key() format.
- Add platform isolation regression test for the bug in #12542.
- Drop decorative test_legacy_key_collision_bug (the fix makes the
collision impossible; the test mutated a single key twice, not a
real scenario).
- Adapter mocks in _sync_voice_mode_state_to_adapter tests now set
adapter.platform = Platform.* (required by new isinstance check).
