Clarify comment in `validChains` in ChainDB q-s-m model

> I agree that this comment is confusing; in particular, it seems very unlikely that we ever have equally preferable chains as we use (non-cryptographic, but still) hashes as tie breakers. This comment was introduced in https://github.com/input-output-hk/ouroboros-network/pull/892 (August 2019), but hash tie breakers were only introduced in https://github.com/input-output-hk/ouroboros-network/pull/2228/commits/13743aedff4e9c4a5c571091cee01ef54f16d334 (June 2020). Hence, I think we could even just remove this comment and the sorting, or, less drastically, add a comment that this should basically never be relevant.
>
> In any case, I don't think we have to change this in this PR (this comment is much older than any of the PRs that are reverted here, in particular, it was not changed when `pruneKnownInvalid` was introduced), but I think clarifying this is a useful separate change.

_Originally posted by @amesgen in https://github.com/input-output-hk/ouroboros-consensus/pull/284#discussion_r1294827186_