Validation of channel keys for time series annotation

[pmayostendorp]

Is your feature request related to a problem? Please describe.
When annotating time series from .csv files, the keys entered into the annotation template must match the csv headers exactly. Literally a single misspelled value in the column tag for the Channel element results in the task failing to properly load data. A blank screen like this will result (I'm using v1.13.1 here, but this is an issue in previous versions as well):

Even the console errors provided are not particularly useful. They are mostly red herrings that resulted in lots of tail chasing for me. For all I know, some other bug reports related to time series could be the result of a simple misspelled key. This issue comes to mind.

Describe the solution you'd like
When opening a time series task, there should be a validation step to ensure that the csv headers requested are available in the time series data source. If they are not present, the front end should display an error indicating that the channels requested by the annotation template were not available in the data source. The user should be prompted to examine their data source and their annotation template to ensure they are compatible.

Describe alternatives you've considered
Probably some additional documentation around setting up time series templates and ensuring that the user understands that just because the template looks reasonable on the labeling interface settings config page does not mean that data will correctly load in a task.