I am sorry to hear that you are running into this issue.
It is actually expected behavior that you won't be able to see the columns there if you are using Coded Step. You can verify the changes selecting the test and clicking Bind Test in the Data Sources ribbon. This will not be a problem if you make sure that the name of the first column is "Geotech (Orange)".
As you can read in this article, you can experience problems changing the CSV file if you are not editing the file in the right location:
Note: When adding an external data file to your project, a copy of that file is placed in your project's Data folder. This is from where the file is accessed and any changes should be made to the file in this location.
The default path in Windows 7 is:
C:\Users\User\Documents\Test Studio Projects\Project Name\Data
I have recorded a short video to demonstrate you can fix your problem (in the video I am not using coded steps, but it will be similar in your case) . You can see on the video that right-clicking on the test file you can select the option "Remove Data Binding" to remove the Data Binding.
About your second question about the image verification:
In general, this is not the type of testing for which Test Studio is designed: we are focused on functional testing of the business logic contained in your application. There are several reasons for this. One of them is the inherent frailty of most user interface testing, your scenario included. That is why we recommend using Image Verification only as a last resort (I would recommend you to take a look at this article). Even if you choose this approach - Image Verification, we would recommend you to compare manually the expected and the actual image.
You can do the verification also with coded steps.
You can capture an image of the browser at any point in the test and save that image to the hard drive. Here's how it looks in code:
System.Drawing.Bitmap image = ActiveBrowser.Capture();
Some other ways of capturing the browser state are listed in this article
After that if you choose to compare the two images with code you can follow the approach from this article
Looking forward to hearing from you.