Re: Floating point comparison inconsistencies of the geometric types

My progress so far is attached as 2 patches.  First one introduces an
header file for adt/float.c.  Second one refactors the geometric
operations.

I have removed the fuzzy comparison macros all together.  It is very
hard to keep some of them and maintain consistency.  Many operators
that would benefit from the fuzzy comparison is quite broken at the
moment [1], anyway.  I have also made the comparisons safe against NaN
[2].

[1]
https://www.postgresql.org/message-id/flat/CAE2gYzw_-z%3DV2kh8QqFjenu%3D8MJXzOP44wRW%3DAzzeamrmTT1%3DQ%40mail.gmail.com
[2] https://www.postgresql.org/message-id/flat/28685.1468246504%40sss.pgh.pa.us