Unfortunately for you, the flat file source expects all records in the file to have the same structure. So, if you have a different number of fields in your trailer, you have a problem.

One possible solution is to split up the file into separate files based on the type of data in each record - use the flat file import but treat each record as a single field and do a conditional split based on the record type. Write the output from each output from the conditional split to a separate file. Now you can import each of the resultant files using separate connection managers which are configured to have the correct number of columns for each of the records.