Fix memoization thread safety

[mkdynamic]

I haven’t tested this (edited this PR on my phone) but in theory this should fix the thread safety issue.

[ScotterC]

Thanks for the contribution!
I haven't had much experience with writing a spec for thread safety issues. Have a sense of the outline you'd approach it with?

[mkdynamic]

Unit testing reliably this would very tricky. IMO it’s not worth it, and you’d mainly end up testing the Ruby Mutex works, which seems pointless.

This pattern of using a mutex to synchronize a memoized class variable in Ruby is common enough — I would say if the existing test suite passes, call it good.

[IAPark]

Could you run rake standard:fix?

I'm guessing we could do this with less locking, maybe eager looking in the cache before locking, but probably safest to just use a mutex for the whole thing

As to testing I agree that there aren't really great options. I think I've mostly heard of people using locks under the hood to make sure the race condition happens, but I think I'm comfortable if all the specs pass with this.

[andreibondarev]

Relevant issue and comment: patterns-ai-core/langchainrb#446 (comment)