feat(hog): fix elements matching

[mariusandra]

Problem

It started from this:

This exposes two problems:

• we don't have the new materialized elements_chain_* columns in Hog globals
• we can't run functions that use lambdas, like arrayExists (solved)

Changes

• Update the "expected volume" calculation to take into account global properties (those not on any one series)
• Remove some unused code from the plugin server (ParsedClichouseEvent was not used anywhere 🤷)
• Add the 4 new fields elements_chain_href, elements_chain_texts, elements_chain_ids, elements_chain_elements into Hog filter matching.
• These fields are stored in the database as materialized columns. We use the same regex that's used by ClickHouse to calculate them in the plugin server.
• The fields are added lazily, meaning they're calculated only when accessed, or when the bytecode is serialized.
• Adds functions indexOf, position, positionCaseInsensitive and arrayCount

Extra

• I'd like to add "elements_chain" higher up on the HogFunctionInvocationGlobals field event, in order to better match the actual database schema. However since it has more discrepancies (name instead of event for the actual event, etc), I decided to leave it for now.

Alternatives considered

• Always calculating these materialized columns into event instead of lazily getting them. Rejected as this will be wasteful if you have no functions that look for element texts
• Moving the regex calculation into Hog. This is a lot more work as Hog doesn't (yet) support lazy properties, and I'd then have to inject some "global hog" code before each filter.

How did you test this code?

Added a test in hog-executor.test.ts, poked around in the UI.

[github-actions]

Size Change: 0 B

Total Size: 1.12 MB

ℹ️ View Unchanged

Filename	Size
frontend/dist/toolbar.js	1.12 MB

_{compressed-size-action}

[posthog-bot]

📸 UI snapshots have been updated

2 snapshot changes in total. 0 added, 2 modified, 0 deleted:

• chromium: 0 added, 2 modified, 0 deleted (wasn't pushed!)
• webkit: 0 added, 0 modified, 0 deleted

Triggered by this commit.

👉 Review this PR's diff of snapshots.

[benjackwhite]

Love this solution. I'm wondering though if we should also cache these values? Maybe not worth it for now...

Also a comment here would be helpful for future travellers to understand what this is all for (I barely get it right now)

[benjackwhite]

Also its not awesome that there is only a test for one path...

[mariusandra]

I think I found a rather neat way to cache this: e2e1bc7

More tests coming soon...

[benjackwhite]

wooohhhhhh self destructing function :o

[mariusandra]

Ready for a re-review