Removed rstrip() for proper line blanking

[DoktorJ]

The addition of .rstrip() to the return output broke line blanking when overwriting the same screen space; removed it to restore previous behavior.

[kroitor]

This is strange, but it breaks the tests somehow. I will investigate and will get back to you with my findingds. In the meantime, if you could check the error here https://travis-ci.org/github/kroitor/asciichart/builds/698977078?utm_source=github_status&utm_medium=notification it would help speed up the process. Thx!

[DoktorJ]

Sorry, I'm just now getting to this; I apparently had a mail filtering issue and never saw the notifications...

It looks like the test is failing because the "expected" graph has no whitespace while the actual graph does (which is the point of removing .rstrip(). I unfortunately do not understand enough about the CI tests to grok how the original PR that added .rstrip() didn't fail its test (since to pass tests prior to that PR, the whitespace would have to have been expected, but would then be absent after the inclusion of .rstrip()), but basically it just needs to happen in reverse: the test needs to be told to expect trailing whitespace on the rendered graph.