They won't kill the industry - for exactly the reason you give at the end. A problem could arise, however, if you end up in a situation like the UK retail bookmakers - here you have games with profiles so unfriendly that a new player would almost never entertain the games - but because the bulk of the money goes through games like that, the bookies just want more like that. Fine - the current customers like them, but new customers would (i would argue) be put off for life by spinning 30 times at 2 quid a spin without a single win - it's terrible. I'm a player, and i won't play UK bookmaker games other than for research purposes. So your pool of players just gets smaller and smaller.
Luckily, the online casinos don't have this problem and the way they are set up means they probably never will - UNLESS they start to dictate game and maths designs (which i have seen start to happen). Data is brilliant, and they have a lot of data - but equally you shouldn't look at data and say "60% of people like HV games, so all our games should be HV". That's a one way ticket to the bottom. Games, profiles, themes, etc.. should cater to as many people as you can. Some of those games might not be top 20 games, but if they cater to a group of players that don't play anything else, then the income is still incremental. You can't just look at the top 10 games and say "we need more of those" without also looking at other metrics or KPI's.
Why would a casino remove stakes lower than 50p? What possible reason do they have to do that? Land-based need high coin-in, and therefore stakes play a part - but online, stakes are much less relevant - you want people playing. There is no occupancy issue - no bums on seats blocking other players - so if someone wants to spin at 10p a spin for 5 hours a day, so what. And also, i believe changing RTP based on stake is something the UKGC don't like...