Reverts actualbudget/actual#1204

Closes https://github.com/actualbudget/actual/issues/1293

This is way more complicated than originally anticipated. Reverting the
change to fix the prod build. Someone can work on re-introducing this in
a later build, but we need to be sure it doesn't break once more.

Bringing back asset caching. This means we won't need to do a full
re-build for every change.

I had previously disabled caching because the backend changes (kcab)
were not picked up properly. But now that should be fixed too.

For #1194

I noticed that:

- The web and electron versions of our `uuid` module both looked exactly
the same…
- …and deferred to the `uuid` package…
- …and the async version just called the sync API.

So now we will just use the `uuid` package directly everywhere.

<!-- Thank you for submitting a pull request! Make sure to follow the
instructions to write release notes for your PR — it should only take a
minute or two:
https://github.com/actualbudget/docs#writing-good-release-notes -->

This will reduce install/lockfile size but won’t affect the built bundle
size. Still nice, hopefully!

Another monkeypatch bites the dust.

AFAIK there is no regression in functionality. But please test
yourselves too to double check.

It me it seems the monkey patch is no longer necessary.. At least I'm
not able to reproduce any `escape` key related issues. But LMK what you
think.

The `@reach/listbox` component is used for the custom select component
(for example: for filters).

- New linter rules are now added
- Perhaps the `createPayee` method i've tagged with a disable should be
deleted.
- I have ignored unused methods in Plaid/Mobile as this stuff is still
WIP/experimental?

Co-authored-by: Jed Fox <git@jedfox.com>

actual-server does not need to import the full actual-app/api package.
It can import only the CRDT stuff.. so I'm extracting it into a new
package to reduce the size of actual-server and make the link between
things more transparent.

~based on #1114~

This brings its build process up to date with the rest of the project.
We can now use Node 20 to build successfully.

Closes #918

Draft because it is untested, maybe tonight I will test it! Just wanted
to get some code out cause i had a spare 5 minutes.

The new parser isn't immediately good enough to replace the old parser,
and I sadly lost the time to contribute more! Sorry! If someone else
wants to take maintenance burden of this code, we can not merge this.
Otherwise, it should go the way of the Dodo it seems.

Thanks!

Closes #1044

1. Upgrade all `cross-env` versions to be the same
2. Added `cross-env` to `desktop-electron` - it was used there, but was
missing as a dep

Updates to the latest version of electron and moves the backend-frontend
communication from node-ipc to websockets. This resolves the previous
roadblock regarding `nodeIntegration` .

Done

- Remove node-ipc in favour of websockets. 
- Move file copying out of `preload.js` to avoid importing module `fs`
there
- Bump all electron pacakge versions to the latest
- Added new package for finding open ports as node-ipc is gone
- Tweaked webpack config for above changes


Partially fixes #468

Questions/ Pending:
- Literally every single test fails for me, presumably some issue with
my setup/environment.
- The websocket communication is not using TLS. I'm not sure how to
enable this, or if we even need to as its all local.
- Still need to create the CI for building/deploying but I'm not sure
where start in this regard as i have no exp with it. Presumably we will
need to point the electron auto-updater to the github releases url's. If
people are happy with this PR I will look at adding the CI before its
merged.
- In dev mode only, I have disabled TLS security becuase my docker
container's cert is not signed. I _assume_ this will be true for other
people who spin up the server on thier own hardware. Perhaps I just need
to change my cert to one from letsencrypt or something...

Notes.
I have not touched javascript in eons so my apologies if the commit
trail is a bit fragmented. I tried to keep them fairly contained and
then there is a slightly gnarly final commit fixing all the linter
issues... Please let me know if you want me to squash some commits etc.

I initially tried to move this to web workers the same way the web app
does it but this was unsuccessful. I have found no way to spin up a
worker in one place (frontend/backend) and then pass this worker to the
other. The electron ipc channels don't allow you to directly pass
objects such as workers, everything is cloned/serialised. Passing a port
number so the other end can spin up its own socket works fine.

---------

Co-authored-by: Shazib Hussain <contact@shazib.com>
Co-authored-by: Jed Fox <git@jedfox.com>

Now we just have one spring animation library

We can use hooks now!

Removing unused `prop-types` dependency.

Upgrading `Downshift` dependency

Another batch of `loot-core` migrated.

Hi there, `Peg.js` is unmaintained, so I figure you all would appreciate
if I replaced it with the drop-in replacement of Peggy. This is work I
am breaking out of #918.

Peggy adds new features like source map support that we could use,
although I do not include that in this change-set. It may be useful for
debugging changes to the .pegjs file we have.e

Hi there, 

I try to tackle #798 here. It was suggested to throw this behind a
feature flag, so here it is!

this does its best to import the problem file in #767. 

I am working on this because it would make my work on #918 easier :)

Feel free to set the feature flag to true and try the new importer. The
date parser is not as sophisticated as the one in `node-libofx`, but I
tried 3 different OFX files, one from my bank, one from the mocks, and
one from #767. They all seem to work well enough on that front, but this
is definitely the weak point of the new implementation.

Let me know what you think!

Removing `react-select` and the new autocomplete. This experiment has
failed, so cleaning things up now.

Another batch of files from `loot-core` migrated to TS (loose mode)

This PR converts everything (aside from electron) from CommonJS to ESM.
It is needed to reduce the changes that will happen during the migration
to Typescript (as TS does not play nice with CJS).

Basically:
- rewrite `require()` to `import`
- rewrite `module.exports` to `exports`
- introduce `ts-node` to run importers so we can convert them to TS too

Lastly, sorry for this larg-ish PR, not my preference but when I tried
to reduce its scope, I would end up with mixed commons/esm that was even
more tricky to handle.

<!-- Thank you for submitting a pull request! Make sure to follow the
instructions to write release notes for your PR — it should only take a
minute or two:
https://github.com/actualbudget/docs#writing-good-release-notes -->

Previously, we were using an alpha version of date-fns v2. Now we’re
using the latest stable v2.

Part 1 of the conversion. Mostly renaming js to ts and making sure
things make still sense. Added also handy TS ESLint rules.

In order to support the various .web/.electron/... I ended up adopting
`index.d.ts` as pattern to share type definition. Let me know if that
makes sense for you too. Right now the function type definition is
duplicated, but the solution will be importing from `index.d.ts` and
using `const fn: FnDef = () => ...` that way we can keep all variants in
sync from a single type file.

Such rewrite however is better done in another PR otherwise we risk
confusing git and loosing history (rename + too many changes). Another
thing that might do in the next PR is convert all files to ESModules, as
things get confusing between CJS exports, ESM default/named and TS adds
extra complains.

Seems to build fine, will test later.