Few things:
1. A drive in a USB enclosure returning I/O read errors could be caused by lots of different things -- there is
a lot of "middleman" stuff going on which you don't have visibility into. The USB/SATA bridge could be crapping out, the I/O errors could actually be happening at the USB layer (not the ATA/disk layer), and so on.
2. You getting an I/O error has nothing to do with a drive being slow. How do I rephrase this exactly. You said, in linear order: "I had a USB-based hard disk that gave me I/O read errors. I tried to "repair the sectors" (no details of how you did this). 40% of the drive was bad (no details provided). It was slow because of USB 2.0." The logic here makes virtually no sense. You're connecting dots that aren't meant to be connected.
3. Removing the drive from the USB enclosure, assuming it doesn't void warranty, and hooking it up to a native SATA bus, is a good idea. It's how you go about determining if the drive itself is bad or the USB enclosure is causing problems.
However, you need to understand
how to examine a hard disk to determine if it truly is responsible for the I/O errors you saw. All you did was run some software (we don't know what), and concluded "no bad sectors". I can safely conclude you did not really understand what you were looking at, did not do a thorough analysis with proper tools, and so on. What you should have looked for was the SMART error log and the SMART extended comprehension error log, which could contained situations where the drive returned ABRT or other statuses (which are not necessarily indicators of unreadable LBAs, but could be indicators of other issues). The short of this is: you have tried to do "bad disk analysis" in a somewhat wonky way, and have concluded a bunch of things + posting on the forum without providing any actual evidence.
4. There is no such thing as a "low level format" through end-user utilities. Please don't use this term unless you know what it actually means. On ATA/SATA drives you cannot do this outside of connecting to the drive directly via a serial interface and issuing vendor-proprietary commands. What you did was just a drive erase (writing zeros, presumably, to every LBA). That is not a "low level format".
Did you save SMART attribute data (screenshots, ANYTHING?)
BEFORE you did a full drive erase? You should have. Seeing the attributes before and after matters -- it can help shed light on if your erasing actually repaired/fixed problems (specifically induced LBA remapping).
5. To answer your final question: you cannot reset SMART attributes (values or anything else). They're defined/set in the HPA region of the drive, or a vendor-specific area.
The only thing that is known to potentially (keyword: POTENTIALLY) reset SMART attributes is updating or re-flashing the firmware on the drive itself (and it must be done with the drive connected to a SATA bus natively, not via USB). Most of the time you won't be able to find a firmware for such drives (or if you can, you won't find a firmware update utility); other times the firmware upgrade utility will migrate/merge the existing SMART data during the upgrade, meaning a F/W update won't reset SMART.
If you really want to reset the attributes, then your only choice is the above, or to RMA the drive (the replacement you'll get will likely have its attributes reset,
but not necessarily).
You do not appear to understand how to interpret the SMART data you're looking at. Meaning: you're doing what a lot of people do: using weird disk monitoring software that makes a bunch of assumptions ("OMG A NON-ZERO VALUE MEANS PROBLEMS!!" (this is wrong)) and in turn you're freaking out. Don't feel bad, you're not alone in this regard -- it's very common, and there is a lot of software out there that tries to "interpret SMART" but does it entirely wrong (SMART attributes are almost always meant for pure human analysis/review, software is not intelligent enough to do this kind of analysis -- a human engineer familiar with disks and the exact model of drive you have can do this).
What you SHOULD have done before ANY of the above was post here and ask for help. I'm one of the few here who can review SMART attributes and step you through using intelligent/decent utilities (mainly smartmontools) to determine if your disk is actually busted. But since you've gone ahead and done a bunch of things beforehand, all bets are off. :/