Make whitespace checking more robust

[OisinMoran]

Currently only looks for the literal " " and fails on other whitespace characters like "\xa0" (non-breaking space). This causes extract_words() to not split words on these characters, and also to pick up stray whitespace as words. Python's builtin string method isspace() is perfect for this.

I've never actually written a test before but from looking at some of yours, this might do the trick. I've attached an example document below. Currently len(words) == 32 but from looking at the document you can see it should be 25, and indeed it is when using isspace().

import pdfplumber
filename = "whitespace_test.pdf"
with pdfplumber.open(filename) as pdf:
    first_page = pdf.pages[0]
    words = first_page.extract_words()
    assert(len(words) == 25)

whitespace_test.pdf