Windows CI timeouts with Electron 30

[goosewobbler]

After upgrading to Electron 30 we get timeouts on the E2Es running only on Windows CI. The timeouts don't seem to be linked to a specific test, indicating some type of memory leak.

After a fair bit of investigation, they don't appear to be linked to either electron-builderor Forge. We are currently working around this by downgrading the E2E and example app versions of Electron to 29.x.y only on Windows.

[goosewobbler]

Reopening as it still occurs after #682 was merged. Tried some different approaches in #645, looks like Lerna will fix it (#683)

[goosewobbler]

Settled on Turborepo, removing the debug logging helped but the issue remains. Occasionally it will pass but mostly it just times out. I tried a lot of things, nothing worked, so we're sticking with the Electron downgrade - now applies to Windows only.

[goosewobbler]

Seeing as this doesn't occur on Windows locally, I think this is another instance of actions/runner-images#7320, with the newer Electron versions being just slow enough to trigger timeouts on CI.

New workaround: split the E2Es and run them on separate Windows runners.