[spec] Monkeypatch for `addModule()` has a stale explication of "same-origin worklet"

[pythagoraskitty]

In the spec monkeypatch for addModule(), the following is stated in a note:

"On user preferences error, addModule() will be aborted at an early stage. However, the error will only be exposed to the caller for a same-origin worklet (i.e. where the initiator document’s origin is same-origin with the module script’s origin)."

It used to be the case that a worklet would be cross-origin if created using a sharedStorage.createWorklet() using a cross-origin script, as the script's origin would then be used as the data partition origin for shared storage access. This behavior has been updated in the spec, however, and will soon no longer be the default.

So the language used to explain "same-origin worklet" needs to be fixed.