Hey, I've been lurking this thread and wanted to add a few thoughts that might be useful. I write some code, but I don't think I'll have enough free time to familiarize myself with KoL scripting to contribute soon enough to do much good. However...
I would consider forking Miner.ash. This mining scenario will disappear at the end of Crimbo 2014 in January, and after that all your changes will just be useless legacy code that should to be stripped or commented out. It's just that much more cruft future contributors have to grok when they want to make changes. This should also make debugging a bit easier, since the crimbo mining scenario (Mine5) is simpler than most of the other mines. You could probably strip Miner.ash down to a pretty short, clean script...
That said, getting to the script to recognize the 6 contiguous "promising" squares and differentiating them from cave-ins may be a little bit trickier than it immediately seems. Fortunately, this is a problem that has been solved many times. Just google "boggle solver" and you'll find a lot of examples. The mining problem is just a simplified version of the much harder boggle problem. Thing is, I'm not sure how fast it'll run even stripped down. It's a popular job interview problem because there are a few ways to solve it and some run pretty fast and some run REALLY SLOW. But I have no idea how much overhead comes with the kolmafia script interpreter. Even the efficient algorithms are a bit processor intensive and might be a bit much for a .ash script.
Oh, and I wouldn't even bother trying to write the logic for when the crimbonar effect isn't active. Just have the script buy and use mining oil until crimbonar and oily legs are both active.