FWIW, full nodes have rejected the most-pow chain on several occasions in the past. E.g. in 2010 there was an incident where software errors caused miners to mine a block that minted a billion bitcoins out of thin air. That fork grew to about 75 blocks. There was another incident in 2013 which made it to about 35 blocks.  More recently, at the enforcement of strict-der, nodes started enforcing the new soft-fork rule with 95% hashpower support, and miners not in conformance with the rule (then a hardfork) managed to produce a chain of blocks 5 longer.

By block count, litecoin's chain has more 'work' than Bitcoin, and yet bitcoin nodes don't follow it.

I think that many people, versed in pre-cypherpunk organizational modalities constantly seek to attribute control to some authority in any system. When they look at Bitcoin they see mining and say "ah, here is the parties that are in charge" (or, alternatively and less  often, some software developers).  They're wrong about that: no one is simply "in charge", and that is a big part of the point and value. It also makes some people uncomfortable, because what they want is the old way of doing things with singular authorities which can be influenced in traditional ways... rather than the personal autonomy of cryptographic proof that backs Bitcoin.