I think asking too many questions in one post is not a good fit for MO. -- Why not just ask one of your questions, see whether it is well-received, and if so, maybe ask the next one a couple of days later?

If the answerers "have actually done most of the work", I think it is up to them to publish their results. Of course, usually they should cite your question in the paper. If "the obtained results are far from good enough for them to accept putting their name there", then however it is likely that the results won't be publishable in a reasonable journal anyway.

The question you refer to is merely a question about the inner workings of the mathjobs website. I think questions like this are just too technical to be on-topic on MO. The given question would rather be a good fit for the mathjobs FAQ.

You can bump a question to the top of the list by editing it (respectively, suggesting an edit and wait for approval), by posting an answer or by offering a bounty. Adding a comment won't bump a post. Without points, I think your only option to bump unconditionally is by posting an answer. (In principle you could delete the answer afterwards, but this is of course not good style.).

I flagged it, and also voted to undelete.

Do you mean this one?: mathoverflow.net/questions/252184/…

Isn't this rather an ArXiv question than a question about MathOverflow?

@Todd: I think you can run into trouble with some non-offensive things as well -- e.g. if you violate copyright or disclose secrets.

I think the purpose of limits like this is to limit possibilities of abuse. The time cutoff for changing votes limits e.g. serial un-upvoting.