Bug in check_target_mapping function

[m-kouhsar]

Hi

Thanks for developing this perfect package!
I found the following bug when I tried to use human gtf data release 111 as target_mapping.

In some gtf files in ensembl database there are a lots of duplicate transcript ID. So, if you use these files with many duplicated target_ids as target_mapping, you will get an stack overflow error in R ( C stack usage is too close to the limit). This is because of huge number of duplicated ids and size of formatted_ids in the following warning message in check_target_mapping function:

warning("There is at least one duplicated target ID in the target ", 
              "mapping. Since sleuth prep is not in gene aggregation mode, ", 
              "duplicated entries will not cause errors during preparation, ", 
              "but may cause unexpected behavior when making any plots or ", 
              "tables. Here is a list of all the duplicated target IDs:\n", 
              formatted_ids)

Using head(duplicated_ids) in the paste function to generate formatted_ids might be helpful.

BW,
Morteza