1. You're overthinking the nature of cheating. Flash games have been put through the wringer using memory-editing tools like Cheat Engine. The only truly secure strategy is an expensive one - record, upload, and replay the game on-server. The cheaper alternative is hand-moderated scoreboards. In any case, high scores are not central to the success of most games - they're a payoff that comes at the end of a session and many users will back out before finishing even one session. (Yes your game may differ, but you haven't presented data to back up the assertion.)

2. It's just too early to move to HTML5 for commercial games - no developer ecosystem, monetization methods are entirely DIY, limited user benefit, etc. That will all change, but I would treat a port like this as a demo project.