They removed OS spoofing just recently, and there isn't a mitigation for Raptor, some think meek might help with Raptor, but its very much up in the air.
I haven't kept up with the space much since then, so am unaware if there is more recent work.
In any case, there are valid threat models where you want to mitigate website fingerprinting but aren't necessarily concerned with AS-level adversaries.
In fairness, most of big tech are AS-level adversaries at this point.
Active attack through BGP-hijacking may be partially mitigated, but this isn't really needed for the most pernicious attacks which are interception/injection from a regional entity that's routing to the broader internet (outbound connections).
The same entities can do early transparent encryption termination for outbound connections (to the general web) since they have their own private signing keys tied to root trust CAs (just not the one the valid cert was issued to), and that lets them collect a treasure trove of forensic artifacts to improve their citizen dossier for advertisers/highest-bidder, or inject content that is ephemeral in nature.
Oh I had missed that, thank you btw! Need more of those BGP monitoring systems...
(and they performed an actual live BGP attack (not just simulation), neat)