Fix TokTok.jl

[Ayushk4]

Refer #28

[codecov-io]

Codecov Report

Merging #29 into master will increase coverage by 18.39%.
The diff coverage is 100%.

@@             Coverage Diff             @@
##           master      #29       +/-   ##
===========================================
+ Coverage   75.88%   94.27%   +18.39%     
===========================================
  Files          10       10               
  Lines         651      524      -127     
===========================================
  Hits          494      494               
+ Misses        157       30      -127

Impacted Files	Coverage Δ
src/words/TokTok.jl	100% <100%> (+19.48%)	⬆️
src/sentences/sentence_splitting.jl	98.11% <0%> (+9.97%)	⬆️
src/words/nltk_word.jl	100% <0%> (+10%)	⬆️
src/words/fast.jl	98.14% <0%> (+16.32%)	⬆️
src/words/tweet_tokenizer.jl	91.15% <0%> (+19.11%)	⬆️
src/words/sedbased.jl	100% <0%> (+84.21%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update f910c2c...00cdf78. Read the comment docs.

[Ayushk4]

@oxinabox I believe this is done from my side, pls review.

[Ayushk4]

Here is one small difference b/w 1.1 and 1.0 that caused error in one of the newly added test cases. (Now fixed)

1.1

julia> collect.(Iterators.flatten(["–—", ("'", "’")]))
4-element Array{Array{Char,N} where N,1}:
 '–'   
 '—'   
 ['\'']
 ['’'] 

1.0

julia> collect.(Iterators.flatten(["–—", ("'", "’")]))
4-element Array{Array{Char,1},1}:
 ['–'] 
 ['—'] 
 ['\'']
 ['’'] 

[oxinabox]

LGTM, thanks