[BUG] normalise should fold case before stop words filtering

[seanv507]

Thanks for taking the time to fill out this bug report!

Describe the bug
stopword_filter is case sensitive return [token for token in tokens if token not in stop_words]
src/utils.py line 58

this means that stopwords at the beginning of sentences will not be recognised

To Reproduce
Steps to reproduce the behavior:

from src.normalizing import normalize, NLTKTokenizer
tkn = NLTKTokenizer()
test1 = normalize("This is a test", tkn)
test2 = normalize("Is this is a test", tkn)
print(test1[1], test2[1])
# outputs['this', 'test'] ['is', 'test']

Expected behavior
'this' and 'is' are stop words and should be removed regardless of whether they are in capitals or not.

Screenshots
If applicable, add screenshots to help explain your problem.