Geometry_Engine: Null handling

[rolyhudson]

trafficstars

NOTE: Depends on

Issues addressed by this PR

Closes #2487

Test files

Changelog

Additional comments

[rolyhudson]

@BHoMBot check null-handling

[bhombot-ci[bot]]

@rolyhudson to confirm, the following checks are now queued:

• null-handling

[rolyhudson]

First attempt at diagramming what is going on with this PR in its current state. Is this insane? Would be good to get a wider consensus on the approach @peterjamesnugent @IsakNaslundBh @FraserGreenroyd @al-fisher.

[peterjamesnugent]

@rolyhudson the diagram looks great- I think it aligns with what we discussed. The challenge is to either make the method generic enough that it can just accept an IGeometry and loop through as you've illustrated with the deep checks etc. or just having IsNull methods for the various objects in the Geometry_oM.

The only thing it doesn't capture is how the messaging is propagated and how much detail that we want to feed back up the chain. Last time we concluded that nested methods shouldn't be a problem because the null will get picked up early, with the deep check at least so I guess there's a decision to be made on how much you want to feedback to the user that it's useful.

[FraserGreenroyd]

This PR has gone stale per our guidelines and following discussion with @IsakNaslundBh it has been decided that this PR would not be suitable given the conflicts, and changes to the Geometry Engine since this PR has been raised. Due to the introduction of check-null-handling it is also difficult to make the situation worse going forward, so although this PR aimed to make things better, we will tackle it as smaller and more discrete tasks going forward. Thanks for the help though @rolyhudson 😄