Handle final periods

[Ayushk4]

Before -

julia> toktok_tokenize("This is a sentence. ")
4-element Array{String,1}:
 "This"     
 "is"       
 "a"        
 "sentence."

Now -

julia> toktok_tokenize("This is a sentence. ")
5-element Array{String,1}:
 "This"    
 "is"      
 "a"       
 "sentence"
 "."

Also, minor changes in handle_final_periods function, to prevent re-traversing over trailing spaces at the end of the string.

[oxinabox]

Can you check this against the original toktok, or against the nltk toktok?
Either way I am infavor but if we are dieviating we should document that this is an enhanced version of the toktok tokenizer

[Ayushk4]

Nltk's toktok gives the following output for toktok.tokenize("This is a sentence. ")

 ['This', 'is', 'a', 'sentence.']

[oxinabox]

Ok cool.
Lets add to the docstring that this is an enhanced version of the orginal toktok tokenzier