Install Plugin Step: Plugin Server side

[neilkakkar]

Changes

Part of #405

So, this turned out to be a lot more involved than I imagined. Have decided to break it down into steps.

This is step 1: Create a new worker pool, with it's own vm/setup/everything.

PR is a proof of concept, very rough, need to clean everything up (and repeat myself less) - but wanted to get this out so you can see what's going on and comment on the approach.

For installation, our setup makes things such that there's only 1 plugin on the hub, which allows us to freely re-use other tasks as they stand. (as long as we don't mess up on reload - I'm thinking of banning reload on this thread).

Everytime a new task comes in, the install thread sets up the plugin, and returns if there's no problems. I've set things up so that this will be re-usable for #165 -> basically every time you want to test something, it begins with a fresh installation, running setup -> task -> teardown.

Some repetition is inevitable, but I think I've whittled it down to the minimum (except vm.ts, which will go away).

Also, naming is very much a WIP too

---

To investigate:

1. Troubles with multiple pools
   2. Pools missing logs
2. DRY

Checklist

• Updated Settings section in README.md, if settings are affected
• Jest tests

[neilkakkar]

How I'm thinking of splitting the steps:

1. Create celery task on Plugin server to accept install commands. (this PR)
2. Update main PostHog to send install tasks on install (not upgrades - those come later, add on). Figure what to do on frontend when install fails (and how to show it)
3. Update schemas / introduce Installation upgrades and testing with different source archives. (see #405 for more)