Language is not essential for the cognitive processes that underlie thought

The claims here are exceptionally limited. You don't need spoken language to do well on cognitive tests, but that has never been a subject of debate. Obviously the deaf get on fine without spoken language. People suffering from aphasia, but still capable of communication via other mechanisms, still do well on cognitive tests. Brain scans show you can do sudoku without increasing bloodflow to language regions.

This kind of stuff has never really been in debate. You can teach plenty of animals to do fine on all sorts of cognitive tasks. There's never been a claim that language holds dominion over all forms of cognition in totality.

But if you want to discuss the themes present in Proust, you're going to be hard pressed to do so without something resembling language. This is self-evident. You cannot ask questions or give answers for subjects you lack the facilities to describe.

tl;dr: Language's purpose is thought, not all thoughts require language

Language's purpose - why it arose - is more likely communication, primarily external communication. The benefit of using language to communicate with yourself via "inner voice" - think in terms of words - seems a secondary benefit, especially considering that less than 50% of people report doing this.

But certainly language, especially when using a large vocabulary of abstract and specialist concepts, does boost cognitive abilities - maybe essentially through "chunking", using words as "thought macros", and boosting what we're able to do with our limited 7+/- item working memory.

For one, how would you know? It left no fossils, nor do we have any other kind of record from that time.

For another, the very question implies a teleological view of evolution, which is arguably wrong.

As for what 50% of people report (where did that number come from?), we have virtually zero intuitive insight into the inner workings of our minds in general, or of the way we process language. All the knowledge that has been obtained about how language works--linguistics--has been obtained by external observation of a black box. (FMRIs and the like provide a little insight inside that black box, but only at the most general level--and again, that's not intuition.)