Better support for BLE devices with less frequent advertisements.

[prestomation]

I wanted to say thank you so much for the amazing integration. The model you have created that automatically uses BT devices, BT proxies, areas, Private BLE Integration/etc is such great software and was soo easy to setup. I had been avoiding room presence for so long as I was fearing the exponential increase in maintenance but bermuda plus ESPHome bt proxies made this so great.

Is your feature request related to a problem? Please describe.
I have some BLE devices that aren't 'seen' as often as others.
I have a Fi dog collar. It is bluetooth/wifi/LTE. It pairs with a base station or an owners phone most of the time for least power consumption. I have this tracked in bermuda.
My goal is room tracking for my dog, so I know where she is in HA.

I did some experimenting. If I powered down the base station and disabled BT on the phones, it works great!
but the area would go 'unknown' after turning on any of these. Not sure of the details but whatever advertisements this thing sends out seem to stop/slow down.
I thought they stopped altogether, but I noticed later that I see rare pings, see image

in this one evening sample the collar went as long as 90 minutes without an advertisement.

Describe the solution you'd like
I'm not 100% sure what I want but wanted to start this issue to try to identify this general problem and what broad solutions maybe could be implemented to improve usability of tracking these devices.
In my case the last known area is better than nothing. maybe ' last area seen' and 'last time seen' sensors? Maybe a per device 'area timeout' I can set to 90 minutes? I also use this integration for the collar so I have a web-powered device tracker for the same device. Maybe something that helps these integrations work together so the area doesn't timeout for longer if the associate device tracker is 'home'?

Describe alternatives you've considered
I have a similar problem with my HA beacon on an android phone. with the default settings I saw similar behavior(but with shorter 'dead periods'. Fortunately HA let's me fiddle with the settings, but this could be useful for this case too.

Additional context
Add any other context or screenshots about the feature request here.