perf: small fix for the insert-method of pgvector

[ArneJanning]

Please describe the purpose of this pull request.
Thankfully @sarahwooders provided a fix for #988 , but assumed a fixed chunk size of 1,000 for pg8000 which slows down the ingestion-process significantly and is not future-proof.

So I added a small method to get the optimal chunk size based on the number of columns in the database.

How to test
memgpt load directory --name <some_name> --input-files=<very_large_input_files>

Have you tested this PR?
I tested it with several very large PDFs and TXTs (around 1.000 pages of scientific content).

Related issues or PRs
#988

Is your PR over 500 lines of code?
No.

[sarahwooders]

Thanks for adding this!

A minor nit - could you please set a self.insert_chunk_size = self.get_optimal_chunk_size in the class initialization, so we can avoid calling the function on every insert?

Also, would you be able to write a test to ensure that the chunking is working properly? Unfortunately all our tests user very small data, so I dont think this functionality will be covered. Maybe you could generate a long list of Passage objects and insert them?

Also, could you please just run the formatter with poetry run black --check . -l 140?