How do researchers send unsolicited emails asking for feedback on their works?

The real answer is that they don't. I'm sorry, but I can't slice it any other way. You would need all the planets to align in order to receive an answer after such an email. You are asking someone who doesn't know you and who is probably very busy (like all academics) to provide you a big favor for nothing in return. Because let's be honest, giving feedback on someone else's work is taxing. The best case scenario is that you get very vague and superficial feedback about the first page of your paper.

Instead, ask your colleagues, or people you have already built a rapport with (an advisor or former advisor, a collaborator...). Or do it the hard way and submit it for publication. You will get (anonymous) feedback. But make sure first that your paper is close to being ready for publication, because you don't want a desk rejection, and you don't want to be known as that person who wastes referees' time by submitting drafts.

Essentially, you have three challenges that work against you receiving an answer to this mail:

1. You need to avoid your mail to appear like academic spam. Your email draft is rather good in that sense because it is short, nicely written, and free of the usual keywords that trigger people's mental spam filter ("submit your article", "endorse this and that", and so on). However, adding a tracking mechanism is certainly not a good idea for this reason alone.
2. You need to avoid appearing like a quack. Both, not pointing to an actual paper and not currently working at a university work against you in that regard. While I agree with Dan that it "should" not matter in principle, I am afraid in practice it will. I assume most recipients are considerably more likely to open an email sent from a person working at a reputable university pointing at an arxiv link or journal paper than a mail from a person without affiliation sending them a link to their personal blog.
3. It needs to be clear enough why it should be a priority for the recipient to read what you sent them. Quite frankly, something being interesting to my work is not enough - my to-read list currently contains some 50 or so papers, all of which are presumably interesting to my work (and some of which I need to review). There just never is enough time to actually read all of them. I think your chances of moving to the top of this list are higher if (a) there is a clear question (that is interesting to me) in the email, and (b) the material is short enough that I can read and comment within half an hour or less.

Hello Dr. X To introduce myself, I am Y, in institute Z.

Looks good so far.

I'm sorry if this email is not convenient at this time, but my work about A is highly relevant to your work in B, so I think it will be interesting to you too.

Why are you “sorry”? I think it’s a mistake to apologize or use words that suggest you feel you are doing something wrong for sending the email. Sending the email is fine, and conversely, if you really were doing something wrong, then you shouldn’t send it.

If possible, can you take a look at it and tell me what you think?

This seems rude to me and decreases the chance (which admittedly is in any case quite small) that you will receive meaningful feedback. This person doesn’t owe you anything, so don’t ask them to do something: they will either do it because they are naturally inclined to do it, or they won’t. I would say instead “I welcome any comments or feedback you might have.”

Also, if you must ask a question, “what do you think?” is a bad question to ask since it is completely open-ended. Make the question precise and specific.

Here is the link: Human interaction with cats Thank you for your reading, hope you enjoy it.

This part looks good.

I wonder if the summary/abstract should be included in the email too, since the principle for asking good question is to show everything you know about it. Maybe it's not necessary, since the title of the link should prove its interestingness nevertheless? I'd like to have your confirm.

Keep it as short as possible. You already said the work is about A and is related to the recipient’s work on B. I don’t think any more information is necessary. If you really think anything more should be added, write something that’s tailored specifically to the person you’re sending the email to and to their interests instead of a pasted summary.

I'm also interested in the case where Y has no Z, and the link is just a collection of observations posted in a blog, not a full paper with proper citation and literary review. It is possible that the observations may have been covered in the field, but Y isn't aware of that yet.

This shouldn’t matter. Blogs can be just as interesting as finished papers, just include the link and a description of your affiliation if you have one.

I also think that if the link is interesting enough, then putting a tracking method in there is fine too? Even if X finds out, they wouldn't feel insulted either, because they find that the link is indeed interesting.

If I noticed a tracking mechanism in the link, I would delete the email immediately, so I would never get to find out if the content was interesting. (There’s also a chance the email would be filtered by various spam filters, and some mail software might preload the link contents so you would think that it was viewed even when it wasn’t. So it’s a bad idea on multiple levels.)