Last halving made it unprofitable for many westerners, but the activity is still profitable in China and the few countries with very cheap electricity. But we are still marching there. If bitcoin price "doubles" as you say, it would only delay the inevitable, only a few more years.
This isn't the case, the "inevitable" can be delayed to as far as a hundred years, it really isn't a matter of opinion or a thought, it's pure math.
Let's run the math for 13 and 33 years from now
Year 2020, bitcoin block reward = 6.25 > total reward in fiat = $62500 when price = $10,000
Year 2032 , bitcoin block reward = 0.78125> total reward in fiat = $62500 when price = $80,000
Year 2053 , bitcoin block reward = 0.0.0244141> total reward in fiat = $62500 when price = $2,559,996
Can bitcoin reach 80k in 13 years from now? very likely, can it get to 2.5M in 2053? maybe!
Mining gear manufacturers end only when at least one of two things happens.
1- They can't improve their mining gears efficiency
This takes away any incentives for miners to switch their mining gears for more profitable once, the network will be fully saturated and miners will start looking for a way to reduce their power cost instead of buying more efficient gears, those who fail to keep up with the low prices will have to leave the game and sell their miners to those who have cheaper power rates, all of this harms the manufacturers and they will be forced to close
2- Mining rewards in fiat become too small
The point which you have explained already which depend ENTIRELY on the price of bitcoin and not the block rewards if we were to reach consensus to increase total supply and increased block rewards to 50
BTC does that necessarily mean mining will be more profitable? no, because if bitcoin drops to a $100 the total mining industry will only generate $5000 a block, which is too small for everybody to sustain their businesses.
If bitcoin price drops to $1000 it will hurt gear manufacturers much more than the halving, so there is nothing that is going to happen in "a few more years" if the price keeps going up, I see that the end of those manufacturers will be the first point, and agree to what you mentioned here
Whatever chip is being used in the S19 might be the last, or perhaps there could be one more from that company
This is because technology is the limit, bitcoin isn't the limit, if the price keeps going up then the network will be able to accommodate more miners, and all they have to do is "make more gears of the same chip".
If bitcoin price doesn't increase, every halving will negatively affect commercial mining which will have a huge impact on the manufacturers.
To sum up:
You think industrial mining will die in a few years because of the halving, in other words, you are betting against the price of bitcoin
I think the opposite, I am betting on the price of bitcoin to go to much higher prices making the halvings near irrelevant for a few decades.