Change check for new points added near but not at an existing point

[jteresco]

From https://forum.travelmapping.net/index.php?topic=7771.0

Introduction of a new point that would become an NMP with an existing point. This might indicate that a waypoint was added to a new route to represent an intersection with an existing route but the existing route's coordinates were not used by the new route or updated to match the new route's waypoint location.

[jteresco]

Some quick thoughts:

• We might want a wider tolerance for detection than the usual NMP detection in the site update process. This would capture more potential problems that could be ignored if needed here and would never become NMPs in the data. The default tolerance is 0.0005 and is not a distance but a difference in degrees latitude and longitude as computed by

DataProcessing/siteupdate/cplusplus/classes/Waypoint/Waypoint.cpp

Lines 101 to 105 in 8c8c808

	bool Waypoint::nearby(Waypoint *other, double tolerance)
	{ /* return if this waypoint's coordinates are within the given
	tolerance (in degrees) of the other */
	return fabs(lat - other->lat) < tolerance && fabs(lng - other->lng) < tolerance;
	}

• It should honor the "LOOKS INTENTIONAL" we have for current NMPs where points are slightly misaligned (by 0.000001) when the point is moved because it is nearby but should not be considered the same point.
• It would be helpful if this could also report the entry for the NMP false positive list so FPs could be added before the changes get into the repository. Even if not, the existence of the check will warn the highway data manager that they should be ready to do so.
• The NMP detection in the current site update code relies heavily on the quadtree for efficiency.