566 fix cjk characters extraction

[wind-chh]

Pull request

This PR try to fix issue (#566 ) and the same failure(but different cause) on my own pdf files.

How Has This Been Tested?

I've added two pdf files under sample/contrib, one is from issue reporter and
the other is my own which occur the same issue. And I've also added two tests
in test_highlevel_extracttext.py testsuite. All unittests passed on my environment.

Checklist

• I have added tests that prove my fix is effective or that my feature
  works
• I have added docstrings to newly created methods and classes
• I have optimized the code at least one time after creating the initial
  version
• I have updated the README.md or I am verified that this
  is not necessary
• I have updated the readthedocs documentation or I
  verified that this is not necessary
• I have added a consice human-readable description of the change to
  CHANGELOG.md

[pietermarsman]

Your changes look good, and I'm happy that all the tests are working. However, I'm not very familiar with this part of the code. Can you indicate how certain you are that this code will not break other PDF's?

Also, I've left some requests for small changes and some questions. If you change/answer them this PR is good to go.