bugs

[behinger]

• [x] if Condition/level is an int, I get an error later
• [x] if #betas or #thetas doesnt fit, I get a hard to parse error
• [x] convert generates a "cond" not a "stimType" column

[behinger]

• [x] multiple subjects can have last / first trial overlapping with eachother - we need to prolong the "onset"-onset (aka first onset) of each subject to fix this

[behinger]

fixed in #84