pandoc-placetable VS pandoc-types

You could do this with a pandoc filter. That's probably more flexible than a pandoc built-in. You can easily adapt a filter to various situations, file types, and so on. After a very limited search on the internet, I do find quite some filters that convert csv formatted data in code blocks to tables. For example, https://github.com/mb21/pandoc-placetable. I'm sure there are also filters out there that take a pandoc link to a CSV file and convert it to a table. If not, they're not difficult to write either, and then you can get exactly what you want.

If you really want to stop using Markdown to write with, then the best solution will be to use a proper conversion tool to turn these into word processing documents, such as DOCX or ODT, and then import that into Scrivener. I don't think (without plugins anyway) that Obsidian has any way of making this easier, but a good general purpose tool for this is Pandoc.