Floating point comparisons

[pjanck]

Reading

• here: https://github.com/buildingSMART/IDS/blob/master/Documentation/testcases-property.md#pass-floating-point-numbers-are-compared-with-a-1e-6-tolerance-14
• and here: Testcase "[PASS] Floating point numbers are compared with a 1e-6 tolerance 1-4/4" #78
• and here: Testcase "[PASS] Integer values are checked using type casting 2 & 3/4" #81

I wonder if it is the documentation that needs updating, or the IDS standard itself expanded as proposed in #78? Either way, something should be changed to reflect the intent and remove ambiguity.

[CBenghi]

Related to #78

[atomczak]

Resolved with #304