Bump NetTopologySuite from 2.0.0 to 2.3.0 - #15

Sourced from NetTopologySuite's releases.

v2.3.0

Milestone: 2.3

• Updated JTS compatibility to 1.18.1.
• Add customizable variant of PackedDoubleCoordinateSequence (#482)
• Remove usage of GeometryCollectionEnumerator from IEnumerable implementation (#508)
• Address one issue that prevented the OverlayNG code from kicking in under certain circumstances, even when configured (#527)

v2.2.0

Milestone: 2.2

• Updated JTS compatibility to 1.18.0. Try out the new OverlayNG code!
• Added a way to invalidate the weak reference in PackedCoordinateSequence (#449)
• Added some helpers to make it slightly easier to get new NtsGeometryServices and PrecisionModel instances.
• Improve WKTReader parsing in certain edge cases (#484).
• More work on improving geometry equality comparisons (#447).
• Enable GML3 writing (thanks @​kristofdegrave).
• Improved consistency in accessing non-X/Y/Z/M values in coordinate sequences.
• Copy GeometryFactory instances more faithfully when the only reason for the copy is to use a different SRID value (#437).

v2.1.0

• Updated JTS compatibility to 1.17.0.
  • See JTS' own release notes for details of what's included in this; those will not be reproduced here.
  • Note that the API breaking change included in JTS 1.17.0 was not ported to this release. You may safely cast Polygon.ExteriorRing, any element of Polygon.InteriorRings, and the return value of Polygon.GetInteriorRingN to LinearRing instead.
• Improve xmldoc.
• Fix some packaging issues.
• Add GeometryFactoryEx, which allows callers to force polygon rings to be a certain orientation.
• Unseal NtsGeometryServices in order to allow subclasses to alter the behavior of CreateGeometryFactory.
• Mark a few useful members on QuadEdge as public instead of internal (they are public in JTS).
• GML readers read Z values now.
• CoordinateSequence and Coordinate methods that accept Ordinate enum values now treat values greater than Measure16 the same way they treat all other "Ordinate value is not present in this instance" situations. In the most literal sense, this is technically a behavior breaking change, but only for people who are both casting integer values >32 to Ordinate enum values and actually using them with Coordinate or CoordinateSequence instances with at least 19 dimensions, at least 17 of which are measures.
• STRtree<T>.Remove now has a possibility of working as expected when T is a value type, instead of never removing anything. It now uses the default equality comparer for the type. Reference types still use object reference equality, for backwards compatibility and performance reasons (it's OK to do this for value types, because Remove was essentially a really slow no-op for them anyway).
• The internal Coordinate subclass that we use for everything other than XY / XYZ / XYM / XYZM now handles "missing" ordinate values consistently with how other Coordinate types handle them.
• Copying a Geometry no longer clobbers its GeometryFactory with a brand new one based on the SRID (JTS uses a separate field for this, so this is more faithful).
• More methods on GeometryFactory have been marked virtual, for consistency with JTS.
• Fix an incorrect port: GeometryTransformer.Transform was inappropriately transforming a LinearRing into a plain LineString, which is never better.
• The center of a Bintree Interval is now computed correctly (JTS does not have a method for this; previously, it was the same as the width).
• Add Envelope.Diameter to get the largest possible distance between any two points that both intersect the bounding box.
• Fix an issue with VoronoiDiagramBuilder clipping.
• Start using Microsoft.DotNet.ApiCompat to help us avoid accidentally making API breaking changes.
• In the same spirit of the improvements to WKT number formatting that came with JTS 1.17.0, also work around a very old bug in .NET Framework / earlier versions of .NET Core before 3.0. For details of the bug and its fix in .NET Core 3.0, see: https://devblogs.microsoft.com/dotnet/floating-point-parsing-and-formatting-improvements-in-net-core-3-0/
• Add IEntireCoordinateSequenceFilter. Very similar to ICoordinateSequenceFilter, but it is only called once per CoordinateSequence, to enable some significant performance optimizations. We do not currently take advantage of this in NTS, but we're already looking at opportunities, and we're excited to see what you do with it!

• 8a2a95d Release v2.3
• 0c9dab3 Make IsValidOp work on CoordinateSequence
• 7a54d3b Fix missing OverlayNG related port of UnionNoOpt (#527)
• c838c50 Add a lower-level CoordinateSequence implementation (#514)
• d37d0c2 Geom collection enumerator no longer yields self. (#515)
• 5ad6633 Add MaximumInscribedCircle check for invalid tolerance
• 8fddc34 Add TestInvalidA
• c9261c5 Add OrientationIndex simple failure test
• 491c2b6 Refactoring for DisjointSets
• 994b2fa Add LAB PartitionedUnion
• Additional commits viewable in compare view

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot merge will merge this PR after your CI passes on it
• @dependabot squash and merge will squash and merge this PR after your CI passes on it
• @dependabot cancel merge will cancel a previously requested merge and block automerging
• @dependabot reopen will reopen this PR if it is closed
• @dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
• @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)