* fix(angular-query): forward queryFn errors to the NgZone

* fix(angular-query): forward mutationFn errors to the NgZone

* refactor: remove wrapping try catches

* docs: fix internal link

* Update docs/framework/react/guides/query-invalidation.md

---------

Co-authored-by: Dominik Dorfmeister <office@dorfmeister.cc>

Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>

* fix(examples): Promise returned from invalidateQueries being ignored in optimistic-updates-cache example

* fix(examples): Promise returned from invalidateQueries is ignored in optimistic-updates-ui example

* fix(docs): Promise returned from invalidateQueries is ignored in optimistic-updates documentation

* Update docs/framework/react/guides/optimistic-updates.md

Co-authored-by: Dominik Dorfmeister <office@dorfmeister.cc>

* refactor(docs, examples): standardize queryClient.invalidateQueries syntax in all onSettled callbacks

* ci: apply automated fixes

---------

Co-authored-by: Dominik Dorfmeister <office@dorfmeister.cc>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

* docs: Add Connect RPC to the list of community projects

Connect RPC has a native TanStack Query integration.

https://github.com/connectrpc/connect-query-es

* ci: apply automated fixes

---------

Co-authored-by: Dominik Dorfmeister <office@dorfmeister.cc>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

Updated several mentions of "React Query" to "TanStack Query".
Wrote "formerly known as" instead of using a not all that well known abbreviation to better support the multilingual audience.

…ers (#8771)

* fix(core): make sure that calling setOptions will always notifyListeners

we have an optimized `shouldNotifyListeners` logic internally in the QueryObserver that will make sure we only notify listeners that are interested in a change because they are using a specific key (tracked query properties)

the optimization to force a skip of the notification when updating options because the optimistic result should already contain all the necessary output falls apart in the situation described in the test and in the linked issue #8741.

while the issue itself is a niche edge-case, the fix is to actually remove a bunch of code, so I'm all for it

* chore: fix other adapters

* chore: fix svelte, too

* chore: fix svelte tests