Reduce URL filter length

[richard-cox]

Summary

Fixes #16215

Occurred changes and/or fixed issues

Reduce the length of the filter used in SSP lists two ways

1. use the NOTIN/IN equalities for all user + all system namespace filters.
   • This was previously done but only for filtering out obscure namespaces (Reduce the url length of requests used to support lists #16031)
   • However we were left in the case where the obscure filters were shortened... but the all user / all system filters increased the bloat again
   • Solution is to also use NOTIN/IN for for all user / all system filters
2. reduce duplicates when filtering by obscure namespaces AND user/system namespaces
   • Instead of helper functions creating the filters and then combining them all together (with dupes) create a map with required in/out namespaces and at the end convert that to filters

Technical notes summary

Before - 948 chars

filter=metadata.namespace NOTIN (c-7sqjx,c-7sqjx-p-b8nfr,c-7sqjx-p-vzbhd,p-8nw96,p-tr99n)
filter=metadata.namespace=c-7sqjx,metadata.namespace=c-7sqjx-p-b8nfr,metadata.namespace=c-7sqjx-p-vzbhd,metadata.namespace=cattle-capi-system,metadata.namespace=cattle-fleet-clusters-system,metadata.namespace=cattle-fleet-local-system,metadata.namespace=cattle-fleet-system,metadata.namespace=cattle-global-data,metadata.namespace=cattle-impersonation-system,metadata.namespace=cattle-local-user-passwords,metadata.namespace=cattle-system,metadata.namespace=cattle-turtles-system,metadata.namespace=cattle-ui-plugin-system,metadata.namespace=cluster-fleet-default-c-7sqjx-10ee7a6fee14,metadata.namespace=cluster-fleet-local-local-1a3d67d0a899,metadata.namespace=fleet-default,metadata.namespace=fleet-local,metadata.namespace=kube-node-lease,metadata.namespace=kube-public,metadata.namespace=kube-system,metadata.namespace=p-8nw96,metadata.namespace=p-tr99n

After - 408 chars (the saving will scale bigger the more obscure + system namespaces there are)

filter=metadata.namespace IN (cattle-capi-system,cattle-fleet-clusters-system,cattle-fleet-local-system,cattle-fleet-system,cattle-global-data,cattle-impersonation-system,cattle-local-user-passwords,cattle-system,cattle-turtles-system,cattle-ui-plugin-system,cluster-fleet-default-c-7sqjx-10ee7a6fee14,cluster-fleet-local-local-1a3d67d0a899,fleet-default,fleet-local,kube-node-lease,kube-public,kube-system)

Areas or cases that should be tested

Exactly the same steps as in #16029

Areas which could experience regressions

Also as per #16029 (This changes the way requests that are used to populate resource lists are formulated, so pretty much any namespaces resource's list)

Checklist

• The PR is linked to an issue and the linked issue has a Milestone, or no issue is needed
• The PR has a Milestone
• The PR template has been filled out
• The PR has been self reviewed
• The PR has a reviewer assigned
• The PR has automated tests or clear instructions for manual tests and the linked issue has appropriate QA labels, or tests are not needed
• The PR has reviewed with UX and tested in light and dark mode, or there are no UX changes
• The PR has been reviewed in terms of Accessibility
• The PR has considered, and if applicable tested with, the three Global Roles Admin, Standard User and User Base

[richard-cox]

/backport v2.13.2 release-2.13

[richard-cox]

/backport v2.12.6 release-2.12

[codyrancher]

By any chances is there a plan to move to requests that support bodies?

There's still theoretical clusters where url scaling could become an issue.

[richard-cox]

The long term plan is to have both obscure and system filtering done server side.

rancher/rancher#52858

We might also have a better way to filter system resources soon - #16204