The moment I saw the streaks, I knew they had to be hot pixels. I have had a similar issue from time to time but as I dither, they usually turn into clusters of multi coloured dots. he hot pixel is always in the same place on the image, but if each image is pointed at a slightly different point in the sky, when your software uses star to align all the images, the hot pixels end up in different spots. The fact that the coloured streaks and the faint streaks are all oriented the same way indicate that they originate from the same cause.
One thought that occurs to me, are you using "dithering"? PHD2 has an option that allows dithering in RA only. The streak appear to be oriented east west, so if you are dithering AND dark frames subtraction is not getting rid of your hot pixels, that may be a solution.
Not sure about the Sharpcap "auto dark" feature. Me, I wouldn't use it, you are effectively ceding control of calibration to software. Much better to calibrate properly in pre-processing. I have tried Nebulosity and several other processing programs but have ended up using Pixinsight almost exclusively. Pre-processing and calibration is very easy once you know what to do and gives great results if you have good data. I would strongly recommend Warren Kellers book (see
https://www.bintel.com.au/product/in...v=6cc98ba2045f ) which is excellent for guiding through the pre-process "process".
Malcolm