My ISP assigned my home an IPv6 address, but the net result is that I get captchas and bot checks endlessly. Even a simple grocery order on Walmart's website yields a dozen "Are you a robot" interruptions during a session.
If your ISP uses CGNAT for IPv4, then Walmart could fix the captcha problem by supporting IPv6, where your address is distinct from the bots.
Walmart uses a litany of external services, presumably including real-time threat/bot analytics. For instance AdobeDTM, which does indeed serve via ipv6. It seems possible that IPv6 could be playing a part regardless of the status of the base site. These bot gates aren't at HTTP responses, but are in client interrogations and javascript triggers while interacting with the page.
Many IPv6 addresses could be mapped to a single IPv4 gateway address, causing the trigger.
And this is well known in the industry. The IPv4 world has had enormous mapping and trust ratings and understanding -- coupled with a scarcity that gives range owners or operators a higher incentive to care about what happens on it -- while a lot of people are still completely in the dark about IPv6 and still treat it like some scary unknown.
Indeed, and residential ranges are wholesale blocked from participating in various services, because of abuse through compromised hosts in residential networks.
Budget cloud providers are wholesale blocked from participating in various services, either at thier local edge, or the remote edge, because of abuse through deliberate malicous customers and/or compromised hosts.
I have Google Fiber, and I can't say I get a ton of captchas (other than sites that have them for everyone, e.g. unauthenticated contact forms). The only downside to v6 was I had to get a new router because my old one couldn't route v6 at gigabit speeds (could easily do gigabit symmetric on v4 only, but topped out at 400/400 Mbps on dual-stack).
Back when I had Spectrum (which was Charter in my area pre-merger), their v6 worked fine as well.