fix #876 Updated term.py to add xsd:normalizedString and xsd:token support for Literals

[bhadoo-aditya]

The lexical space of xsd:normalizedString is unconstrained (any valid XML character may be used). Its value space is the set of strings after whitespace replacement—i.e., after any occurrence of #x9 (tab), #xA (linefeed), and #xD (carriage return) have been replaced by an occurrence of #x20 (space) without any whitespace collapsing.
#876

[coveralls]

Coverage decreased (-0.005%) to 75.761% when pulling fedb4f7 on bhadoo-aditya:Sweb into 037ea51 on RDFLib:master.

[edmondchuc]

Merging this closes #876.

[edmondchuc]

Spec reference: https://www.w3.org/TR/xmlschema11-2/#normalizedString

xsd:token is a derived datatype of xsd:normalizedString. Therefore, this PR should also apply to xsd:token as well. See https://www.w3.org/TR/xmlschema11-2/#token

[edmondchuc]

Note that while xsd:normalizedString has whiteSpace = replace, xsd:token instead has whiteSpace = collapse.

[edmondchuc]

Not clear why this needs to be in an try except.

[edmondchuc]

This PR performs normalisation on xsd:normalizedString and xsd:token. For xsd:token, this PR also strips trailing whitespace (left and right) and replaces contiguous whitespace with a single whitespace.

Dockerfile base image version tag bumped from python:3.6 to python:3.7.

Improved some tests for xsd:normalizedString and added tests for xsd:token.