Unlike Tolkien, Lovecraft never really developed R'lyehian into a full language - he didn't need to; he just wanted some creepy sounding gibberish for his stories. This means that getting a word list for "valid" R'lyehian words is not straightforward. I'm not really a Lovecraft fan myself (sorry), so my source for this was mainly this website:
R'lyehian - CthulhuWiki
Notice that a lot of that has been reconstructed by some fans, and it isn't necessarily 'official,' so if Anukan was using a different source for what constitutes Valid R'lyehian then my efforts are already off track.
Notice that there are prefixes, suffixes, and base words. So the first thing I did in creating my word list was to make every possible combination of prefixes, bases, and suffixes. Just to be safe, I let every entry have 0-2 prefixes, one base, and 0-2 suffixes. Although there are only 82 entries, after generating all the combinations the word list contained over 20k entries. Then I wrote a script that created a new list with all the vowels removed. And yes, I added in "fhtng," in addition to "fhtgn" (which showed up in the list already).
So I ran these through the script from last time that tried to match either three words (6, 6, and 10 letters) or four words (6, 6, 5, and 4 letters).
Nothing. No matches, not even nonsensical ones abut nonfat R'lyehian unity-stew.
Next, I tried a few different approaches to the apostrophes. First, I thought that maybe the rune that looks like an apostrophe actually might represent one. R'lyehian is full of apostrophes, after all. So I locked in an apostrophe at character 6 of the last line and ran the scripts again, but this too got no results.
Another approach might be that the characters with two dots, which appear also in un-dotted forms, might mean that it's that letter followed by an apostrophe. In this system, the first set of runes plus the first line of the second set would have the form AB'AC'DE AB'BFGH. The last line could be one long 10-letter word or a five and a four letter word (if the apostro-rune is indeed a word separater).
Again, no matches, so I tried just focusing on the first rune set.
(After all, maybe the two backgrounds translate to text in different languages.)
With vowels included, only two words even fit the form of the first set of runes.
cf'ch'og = c- [prefix: we/our] + f'- [prefix: they/their] + ch' [cross over/travel] + -og [emphatic]
≈ We and they travel!
cf'ch'or = c- [prefix: we/our] + f'- [prefix: they/their] + ch' [cross over/travel] + -or [force from / aspect of]
≈ We and they were forced to cross over (or, smoething? Heck if I know how to tranlsate this).
Neither of these even so much matched on the second set of six letters, let alone the 10-chara line.
Without vowels, five choices fit the first set of runes (again, with this particular interpretation of the apostrophes).
cf'ch'gl apparently from cf'ch'agl (c- + f'- + ch' + -agl [suffix: place)
cf'ch'th apparently from cf'ch'oth (c- + f'- + ch' + -oth [suffix: native of])
cf'ch'gr apparently from cf'ch'ogor (c- + f'- + ch' + -og + -or)
cf'ch'rg apparently from cf'ch'orog (c- + f'- + ch' + -or + -og)
cf'ch'rr apparently from cf'ch'oryar or cf'ch'yaror (c- + f'- + ch' + -or + -yar [suffix: time of /moment] or " ... + -yar + -or")
It's all nonsense! After spending time trying to glean meaning from recombinations of gibberish character strings I began to feel like I was going a little crazy (not unlike many of Lovecraft's characters), so I gave up. I was pretty much out of options, anyway. Although Anukan said we have enough info to brute force a solution, I'm apparently not using the right methods, heheh.
If anyone wants the words lists I created, let me know and I'll figure out how to post them (they're too long to just put in comments). But, I think most of you are not working on this approach anyway. (Good for you - it didn't get me anywhere!)