The [name of db]_Temp is an unintended leftover of the import process.
In this answer Nick links to the stored procedure that takes care of the import.
Here are the most relevant parts:
CREATE PROCEDURE [dbo].[sp_Refresh_Database] @DBName nvarchar(100)
Declare @TempDBName varchar(100) = @DBName + '_Temp';
-- initialization code omitted
Exec('Create Database [' + @TempDBName + ']
ON PRIMARY (NAME = N''' + @DBName + ''', FILENAME = N''' + @DataPath + @BaseFileName + '.mdf'' , SIZE = 4096KB , FILEGROWTH = 102400KB)
LOG ON (NAME = N''' + @LogFileName + ''', FILENAME = N''' + @LogPath + @BaseFileName + '_log.ldf'' , SIZE = 1024KB , FILEGROWTH = 10%)');
-- import of tables code omitted
Exec('Alter Database [' + @TempDBName + '] Modify Name = [' + @DBName + ']');
For unclear reasons the procedure did start and created the Temp db but it never ran to completion. This hypothesis is further confirmed by the fact that the
max(creationdate) of some of the major tables in the [stackexchange.codegolf] database are on April 29th, 2018. That is Sunday a week a go.
It messes up your query because the permissions are not set, so the svc_sede user doesn't have any access to that schema. But the database shouldn't really exist at this point.
An SE developer need to look into the logging to determine why this particular database couldn't be synced from the source and when they fixed the root cause, the procedure should be re-run for [stackexchange.codegolf].
I'm not sure if you want to workaround this issue as it clearly indicate something is off with the SEDE data but in case you do:
where database_id > 5
-- ignore _Temp databases that might be a left over of a broken (or running) import
and name not like '%_temp' collate sql_latin1_general_cp1_ci_ai