feat(db-postgres): optionally allow to store blocks as JSON and add support for "in" operator within richText

[r1tsuu]

Description

This PR adds option to store blocks as JSON column in Postgres with storeBlocksAsJSON property for the adapter.

While i like that you guys choose a hard, traditional way in implementing blocks with SQL, i think this may be not for everyone.

I like the strict approach to a schema when we talk about things like Users, Products, Orders etc.

But when it comes to a sites content it's different, not only that we can have too much things here (resulting into long table / constraints names, many joins on SELECT), it could also be changed frequently and you need to deal with a big migrations.

With Lexical blocks feature - they are stored in the same way, so i guess it's fine to allow that here too?

This PR also adds support for querying with in operator on richText and as well blocks when it's JSON.
This was done in order to pass as many tests as i can with storeBlocksAsJSON. And here's the result for "fields -> blocks"

Yes, not great here with querying on richtext, which is nested to already json blocks.

If this is something that will block this PR to merge, i have an idea on how we can improve overall our JSON queries, it could work for any nested structure. And JSON blocks, blocks in Lexical will work too.
This in theory could be achieved with this query:

CASE
WHEN jsonb_typeof(data) = 'array'  THEN

Why? we don't know in runtime, is current key object or array, because we don't know the structure when dealing with JSON field type. But with this Postgres JSON query way we can determine that.

In any way whatever separated PR or here i want to try implement this, because for now JSON / RichText querying isn't great in Payload.

p.s wrong base branch caused first commit, i reverted it.

• I have read and understand the CONTRIBUTING.md document in this repository.

Type of change

• New feature (non-breaking change which adds functionality)
• This change requires a documentation update

Checklist:

• I have added tests that prove my fix is effective or that my feature works
• Existing test suite passes locally with my changes // with passing storeBlocksAsJSON: true 3 tests are failing.
• I have made corresponding changes to the documentation

[github-actions]

This PR is stale due to lack of activity.

To keep the PR open, please indicate that it is still relevant in a comment below.

[maximseshuk]

When can we expect this PR? This feature is really useful when there are around 50 blocks in one collection. Right now, I'm using a workaround that works great, but I'd love to have native support.