Fix intermittent failure in AjaxBlockTest

I want to RTBC this, since it looks completely simple and straightforward to me, but I'm wondering about that SQLite failure. I know SQLite database locks are a known problem for us, but I'd feel better if someone in the know (looking at you, @xjm) could confirm that those are affecting HEAD from time to time.

I always, always queue only MySQL and Maria environments for testing random fails, because Postgres and SQLite are prone to DB problems. SQLite especially is prone to database locks, and that's what happened here:
SQLSTATE[HY000]: General error: 5 database is locked.Does the database file exist?Does web server have permission to write to the database file?Does the web server have permission to write to the directory the database file should be created in?

So we can ignore that result. However, I'm going to queue a couple more jobs on MySQL environments to be safe.

Okay, then I'm kicking this to RTBC on the assumption that the MySQL jobs will pass.

On the other hand, looks like a reroll of #2 is necessary...

I think it is RTBC for 10.1.x; just we will need a different backport patch for 9.4.x.

Committed to 10.1.x, 10.0.x, and 9.5.x, thanks!

NW for the 9.4.x backport.

Added a patch for Drupal 9.4.x.

I compared the patch in #14 to what was committed. They look identical to me, so I queued up a bunch more MySQL environments. If they come back green, and I'm guessing they will, I'd say this is RTBC.

Committed to 9.4.x. Thanks!