improve puvspr.dat queries

[hotzevzl]

as discussed with Alicia - implementation based on view (which I have kept for the moment) does not allow (as far as I can think...) to prefilter by scenarioId - with the implementation proposed in this PR I have been able to get query times down from ~4min with test data to ~20secs (most of the deal is actually indexes, but still, we also filter out data we won't need)

[vercel]

This pull request is being automatically deployed with Vercel (learn more).
To see the status of your deployments, click below or on the icon next to each commit.

marxan – ./app

🔍 Inspect: https://vercel.com/vizzuality1/marxan/DyksnAJzWiQ9nqEpFXPUDaWBLmu8
✅ Preview: https://marxan-git-chore-apiimprove-puvsprdat-queries-vizzuality1.vercel.app

marxan-storybook – ./app

🔍 Inspect: https://vercel.com/vizzuality1/marxan-storybook/CBXsULxxj6GpWghVPE4PiJVSS3bV
✅ Preview: https://marxan-storybook-git-chore-apiimprove-puvspr-9971ef-vizzuality1.vercel.app

[kgajowy]

PR seems to mix three different ones (deno bot, calculating protection status and improving puvspr dat). Would be great to clean it up first before jumping into review. Thanks!

[kgajowy]

you can just use puvsprRepo.query instead creating new connection.

[hotzevzl]

yep, I know - I did so because with this PR the puvsprRepo should become redundant and we should be able to drop it altogether: I left it in the current PR in order not to change too much in case the new approach ends up not working out in practice, but all going well we should be able to drop the repo.

[hotzevzl]

PR seems to mix three different ones (deno bot, calculating protection status and improving puvspr dat). Would be great to clean it up first before jumping into review. Thanks!

sorry about the PR stack mess - this is properly isolated now 😬