Other than that it's a niche browser with a tiny marketshare and it would increase their workload to first make it work and then keep it working?
1) 1Password supported Arc browser, which is in closed beta, with a miniscule fraction of Orion's userbase 2) Other than an allowlist of sorts so the browser extension can communicate with the Desktop agent, there's no technical reason why the effort would outweigh the benefits (for the users)
For example, a project that made Chrome users 1% better off is probably worth far more to the overall 1pw user base.
(2) No, the cost of supporting a browser is the cost of supporting that browser's users. Even if the entirety of the changes required to enable it is simply adding "com.kagi.orion" or some such to a plist, there's a cost to responding to every "it doesn't work here" message. Arc looks to be VC funded so it seems plausible they could have a commercial agreement with 1password, or it's possible that they raised 17 million to make a chrome wrapper and so have exactly chrome's web extension implementation which would reduce support costs to zero (though I do wish someone would give me 17million for some random project :D )