My answer to your question is a question: what problem would a hover state solve?
Every visual element on a form requires processing by the user's brain. And every additional bit of such cognitive effort increases the burden to the user, heightening the chance of fatigue, errors, confusion and abandonment.
Consequently, the most usable forms are designed by starting with a blank screen and only adding visual elements that help communicate to the user how to use the form. This is why text fields need a border (to communicate "here's the box you type in"), but thanks to our brains being especially wired to notice visual differences, that border doesn't need to be glowing and bright yellow – a simple grey or black one pixel line is enough.
So, if we come back to your question, having a field change its appearance on hover will add to the cognitive burden. Therefore, there needs to be a good reason to do it.
- If you're concerned that people won't know which field to use, locating the field label in close proximity to the field, or using zebra striping, addresses this problem.
- If you're worried that people will not know where they are up to, then you can relax, because this is indicated by all the fields prior being filled, and all the fields following being empty.
- Perhaps you've seen lots of other forms do this? Unfortunately, web/desktop/mobile form design is as much a victim of useless fashion as any other field. And as we all know, fashion isn't necessarily functional, let alone a good idea.
I've tried to think of other reasons but can't – let me know if you do.
I would also support what Jørn and Lie have said: the eye is not necessarily where the mouse is. Given this, the hover state could actually be distracting and interrupt the user.
Finally, don't forget that not everybody uses a mouse.