fix: increased postgres cpu usage after switching to a single get jobs query

[atzoum]

Description

Switching back to multiple queries in router, batchrouter & stash loops, while keeping jobsdb refactoring introduced by #3742.

JobsDB changes:

• getJobsDS optimisations so that efficient joins are used for different stateFilters
• GetToProcess uses 3 distinct queries, as it did before chore(jobsdb): unify get processed and unprocessed queries #3742 changes

Security

• The code changed/added as part of this pull request won't create any security issues with how the software is being used.

[codecov]

Codecov Report

Patch coverage: 79.48% and project coverage change: -0.08% ⚠️

Comparison is base (96588bb) 69.13% compared to head (8c5ba37) 69.06%.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #3812      +/-   ##
==========================================
- Coverage   69.13%   69.06%   -0.08%     
==========================================
  Files         349      349              
  Lines       51947    52023      +76     
==========================================
+ Hits        35915    35928      +13     
- Misses      13730    13784      +54     
- Partials     2302     2311       +9     

Files Changed	Coverage Δ
processor/stash/stash.go	54.57% <66.66%> (+2.05%)	⬆️
jobsdb/jobsdb.go	72.44% <80.28%> (-0.14%)	⬇️
router/batchrouter/handle.go	63.21% <82.35%> (+0.29%)	⬆️
jobsdb/queued_db_request.go	90.00% <100.00%> (ø)
router/internal/jobiterator/jobiterator.go	100.00% <100.00%> (ø)

... and 7 files with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.