Dashboards: Very slow loading after migration to v12

### What happened? After migration to v12, all my dashboard are extremely slow to load (around 80s) I use PG as backend database. I can easily reproduce with a single API call ``` time curl -s 'http://127.0.0.1:3000/secure/api/dashboards/uid/cebd10d9-b6e3-49bd-89a0-547a5ad4a1af' -u `cat ~/.grafanapass` -o /dev/null curl -u `cat ~/.grafanapass` -o /dev/null 0.00s user 0.00s system 0% cpu 1:18.42 total ``` When looking in logs, I can see this request send on PG database ``` Jan 29 11:39:10 intreporting grafana[2592266]: logger=sqlstore.xorm t=2026-01-29T11:39:10.674091948+01:00 level=info msg="[SQL] SELECT dashboard.*, dashboard_provisioning.name, dashboard_provisioning.external_id , dashboard_provisioning.updated as provisioning_updated, dashboard_provisioning.check_sum FROM \"dashboard\" INNER JOIN dashboard_provisioning ON dashboard.id = dashboard_provisioning.dashboard_id WHERE (dashboard_provisioning.dashboard_id = $1) []interface {}{25555} - took: 38.59317ms" caller=[] Jan 29 11:39:10 intreporting grafana[2592266]: logger=sqlstore.xorm t=2026-01-29T11:39:10.675895841+01:00 level=info msg="[SQL] SELECT COUNT(*) AS count FROM dashboard WHERE org_id=$1 AND is_folder=false []interface {}{7693} - took: 37.370003ms" caller=[] Jan 29 11:39:10 intreporting grafana[2592266]: logger=sqlstore.xorm t=2026-01-29T11:39:10.712765692+01:00 level=info msg="[SQL] SELECT dashboard.*, dashboard_provisioning.name, dashboard_provisioning.external_id , dashboard_provisioning.updated as provisioning_updated, dashboard_provisioning.check_sum FROM \"dashboard\" INNER JOIN dashboard_provisioning ON dashboard.id = dashboard_provisioning.dashboard_id WHERE (dashboard_provisioning.dashboard_id = $1) []in…