To be fair, FE is not targeted.. effort is given to all cases. But the developer openly says that preference will be given to easy cases, and harder cases will be worked on where possible. some cases it is stated that there is no solution. This is a good way to get the most cases done possible..
You have to remember that flash vendors do not give out any info on their schemes. It must be reverse engineered. The only way new ECC/XOR/Bad column/page structs/etc etc schemes are worked out is from prior experience and working from there to figure out the new scheme. There are some that none of us have yet worked out.
Don't get NAND and controller chips mixed up and don't assume that if it is xxx controller then it will be xxx scheme.
If it isn't available in the FE library, you can upload the dumps to SC (free support) and quite often you will get a solution. There is no telling which cases will be solved, bit errors, bad dumps can play a part.
I am not sure I like to categorise any tool into "library driven" or "brute force".
VNR is research based I guess and with Tech Partners around the world helping out finding new XORs, plus the access to a huge pool of research drives, there is a great driving force to support more and more. plus the developers of VNR are not new to flash by any means.
If I had to make one recommendation, it would be to start learning about flash.. Pages, Crystals, Planes, CE, blocks, ECC, FAT tables, Files etc.
