There hasn't been enough of a market so they have not made the investment.
Android occupies the lower-end side of the phone market, so there's less of a drive for the absolute highest performance. There has until very recently been almost zero desktop or laptop ARM market or server ARM market.
Apple also has a ridiculous amount of cash sitting around and could afford to fund R&D ahead of market demand and develop a truly killer ARM architecture. Everyone else is now behind, but what I'm saying is that there's not much in the M1 that is secret. It's just well-understood performance CPU engineering techniques deployed well on an ARM chip and the added efficiency is largely due to the lack of X86 cruft overhead and the 5nm node.
The obvious question is why has Qualcomm failed so hard at even keeping up with the A series? They're untouchable for radio stuff but their CPUs are very average. Samsung are also trying very hard with average results.
The M1 isn't Apple's first crack at this that is some fluke which will quickly be overtaken by someone else. Everyone else has already been in a race to beat their A-series chips and failed miserably. The M1 is just the first time we've been able to benchmark it on a real OS. If someone beats it, it will be AMD retooling their cores and doing it in 3-5 years (Zen was originally battling an ARM variant internally at AMD)