So I did test with a number of different fonts yesterday night, and discovered some issues that may make this a somewhat bigger task than simply finding an appropriately licensed replacement that is pleasing to the eye.
First of all, there was still a bug in rendering that caused segfaults when using certain fonts. That took a while to find, but fixed it is (just not yet comitted)
The second, and larger problem is that different fonts have different pixel sizes, when rendered at the same font size. There was already some code in place that would decrease the font size until a certain pixel size was reached, but it did not really work out well. So more time was spent in trying to improve this, without coming to a completely satisfying result.
What I have found on the web on this topic is: "Can you determine the point size of a font by measuring it? No. Because of the differences in apparent sizing between fonts, there’s nothing you can measure that would be conclusive. The only way to figure it out is to set the same text, in the same font, with the same line length. Then adjust the point size so it matches the reference sample."
While I guess I could do it that way (which still sounds dubious, as some fonts are wider or narrower than others), the question is what to do if a font does not contain the glyphs for the sample? Although I guess it might be reasonable to expect that at least standard ASCII is present, I am quite certain it is not always the case.
Btw., leaving aside the matter of size for a moment (which might improve things a little once I have found a way to render each font at a roughly equal pixel size), from all the fonts suggested so far, the only one that looked halfway decent was Aclonica. It seems that fonts with serifs are too thin, so looking at others might not hurt.