Remove setnafill() invalid input error

[deh277]

Closes #7449

Removed an unused error message from the po and pot files so they are less cluttered and more accurate. The error message removed (“‘x’ argument must be numeric type, or list/data.table of numeric types") is no longer being used since #3992 has changed nafill to allow non-numeric types. We made sure to run the test suite on our repository and everything passed on our end.

Worked on by @deh277 and @shantz14

[shrektan]

chinese translation is ok for me

[jangorecki]

Suggested change

	7. setnafill() old non-numeric input error message removed from all .po and .pot files.[#7449](https://github.com/Rdatatable/data.table/issues/7449) This reduces clutter and improves file readability, while keeping the new errors in place. Thanks to @MichaelChirico for the issue and @deh277 for the fix.
	7. setnafill() old non-numeric input error message removed from all .po and .pot files. This reduces clutter and improves file readability, while keeping the new errors in place. Thanks to @MichaelChirico for the issue and @deh277 for the fix. [#7449](https://github.com/Rdatatable/data.table/issues/7449)

[MichaelChirico]

Hm sorry I think there's some misunderstanding. The point of the bug as of now is to try using setnafill() with various invalid inputs, and see if you find the errors acceptable / clear enough.

If so, we can just close the bug.

The .po / .pot files will get updated in time for the next {data.table} release, there's no need for a PR like this (there are dozens of other messages that would need to be updated in a similarly incremental fashion)

[deh277]

Hm sorry I think there's some misunderstanding. The point of the bug as of now is to try using setnafill() with various invalid inputs, and see if you find the errors acceptable / clear enough.

If so, we can just close the bug.

The .po / .pot files will get updated in time for the next {data.table} release, there's no need for a PR like this (there are dozens of other messages that would need to be updated in a similarly incremental fashion)

Hey Michael,
Sorry for the misunderstanding. After some testing we have come to the conclusion that the error messages are sufficient and that the issue can be closed, along with this pull request.