And you're probably underestimating the costs associated with stealth paints / stealth capabilities. Its not easy to avoid modern radar systems. You've pretty much got two choices:

1. Go cheap and swarm -- Ignore stealth entirely, but you pop up on radar like a sore thumb. Which opens up your drones to a variety of automated weaponry. CIWS defense are basically radar-based aimbots that can shoot down supersonic threats like cruise missiles. I find it unlikely that a modern drone would go faster (or cheaper) than a typical rocket lobbed at Israel's Iron Dome, but think of that for a typical defense system you're trying to break these days.

2. Go expensive and stealth -- Stealth capabilities are clearly the future, but their extraordinary costs (F35 project) are well known. Furthermore, stealth capabilities make communication (radio) extremely difficult. If you emit a noise, you can be detected, so a human in the loop (ex: F35) just seems innately more reliable, since they can be "cut off" and enter radio silence for extended periods of time.

--------

Swarming the Israeli Iron Dome with many, many rockets is a strategy that seems to work on occasion. Its not a perfect air defense, so with enough rockets, it begins to miss.

But each of these dumb rockets are launched extremely cheaply: without much computer smarts at all, just flying as fast as possible to make it harder to get shot down. Once we start talking about computer systems (especially in 2021, a year where we've run out of computer chips), your costs escalate severely.

The cheap-and-swarm methodology, especially if you're not really concerned about where you hit exactly (Hamas is mostly doing this as a political / terrorism ploy), is kind of well optimized already. And I'm not entirely sure if an organization like Hamas would switch their rockets to drones in an attempt to beat Israel's Iron Dome. In fact, I'd argue that such a move would be worse than what they're already doing.