Thanks again @Pppery! I just opened a separate feature request, T410208.
I agree with your last comment about the list basically containing the same things, but now - following the introduction of "isDisambiguationPage" - it contains more of them (and even more after T395366 will be resolved).
Thanks a lot again!

@Pppery, thank you very much for the very detailed answer!
I understand, and I agree that it is useful to have a list of all pages affecting a given page; although I'm not sure that the templatelinks table is the best place for that (but I guess that until something like T268526 is implemented, that list can sit in this table, if no better place exists. And thanks for the reference to that task! Also, changing its title to reflect its actual context, which you performed in T406769, is important).

Okay, fair enough, that makes sense.
But it's still not clear to me why the target of a disambig check should be treated as a transcluded page. There's just one bit of information from that page which affects the calling page (disambig or not); that's still quite far from including the entire page contents.

In T401556#11072785, @Pppery wrote:

I disagree with this conceptually; checking whether the page is a disambiguation page is a kind of transclusion. But whatever.

Thanks @Izno!
See my edit at the bottom of the task description.
Could it be that there has been some change in the MediaWiki font settings around that time?
Or is it more probable that the change took place in Chrome?
Is there any way to check?
Thanks again!

In T67550#5213188, @Dvorapa wrote:

In T67550#5213145, @Pols12 wrote:

In T67550#1731373, @eranroz wrote:

As part of T64217 we are going to add "format" which can be either "inline" or "block". Do you think "subst" should also fit to this "format" or do we want it in different field? (preferred format of inline/block doesn't have real meaning when template is substituted)

Indeed, defining a new substitute possible value for format key seems me to be a good interface implementation.

+1

Thanks a lot to both of you!
@Izno - after your comment I found this wishlist survey discussion, perhaps the one you referred to. Just thought to link to it from here.
I agree with @Esanders that this specific feature can be implemented with less side effects by making the visual editor recognize the templates and add the subst: wikicode, instead of assigning this task to the parser.
One small thing I forgot to mention earlier (relevant to all of these suggestions, I guess): in other languages there are local translations for subst:, so this feature should also have some kind of internationalization.

Hi, it seems that the issue exists also for double quotation marks, " (again, combined with Hebrew letters).
Thanks to the user who reported this case, and thanks @Kipod for verifying!

Thanks a lot @Etonkovidova for the detailed image-supported reproduction of the issue!
If I may just add a tiny comment: the same issue exists also for the watchlist. I mentioned it only very briefly in the original report (third-to-last bullet), and this is the case for which @Kipod's comment is relevant.
The interfaces of the two lists (recent changes and watchlist) are very similar, and I assume that the two issues share the same root cause, but I just wanted to mention it for the record.
(For the watchlist, issue (2) described above may occur also not in "extreme cases", for example if the user mostly tracks hidden categories. This is how I originally found out about this whole issue...)
Thanks again!