From what I know from having worked at Nokia, it's simply the result of design by committee where the committee was made up of mutually very hostile hardware (mostly) companies not particularly good at software (that's how Nokia failed in a nutshell), telcos, chipset manufacturers, hardware manufacturers, etc. And I mean hostile as in looking to squeeze each other for patent licensing, competing for the same customers, and suppliers. All this happened in an ecosystem that also produced such monstrosities as gprs, 3G, 4G, etc. Ericsson and Nokia were on top of the market when they created bluetooth and had a relatively big vote in these committees.
Each of those standards were burdened with lots of little features to cater for the needs (perceived or real) of each of the committee members. It's a very similar dynamic to what happened to bluetooth. A lot of 3G stuff never really got any traction. Especially once Apple and Google decided that IP connectivity was the only thing they needed from 3G/4G modems and unceremoniously declined to even bother to support such things as videocalls over 3G. Apple did Facetime instead and in the process also strangled SMS and cut out the middlemen (operators) from the revenue. Google was a bit slower but on Android a lot of 3G features were never really implemented as they were mostly redundant if you had a working internet connection, fully featured SDKs, and a modern web browser.
It's the same for a lot of early bluetooth features. Lots of stuff you can do with it; lots of vendors with half broken implementations with lots of bugs; decades of workarounds for all sorts of widely used buggy implementations; etc. It kind of works but not great and making it work great in the presence of all those not so great products is kind of hard.
Just a long way of saying that bluetooth is so convoluted and complicated is because the people that built it needed it to be that way more badly than they needed for it to be easy to implement (including by others). At this point it's so entrenched that nothing else seems to be able to displace it. I'm not even sure of people actively putting time and resources in even trying to do that. I guess you could but your product just wouldn't work with any phone or laptop in the market. Which if you make e.g. headphones is kind of a non-starter. It's Bluetooth or nothing. I wouldn't be surprised if Apple considered this at some point and then ended up not doing it. They have a history of taking good ideas and then creating proprietary but functional implementations of those ideas.