ST6RI-724 Model library updates from SysML v2 FTF Ballot #9

[seidewitz]

This PR implements resolutions of issues from SysML v2 FTF Ballot 9 that called for updates to library models. In addition, it makes a further revision to TypeAdapter::removeUnnecessaryImplicitGeneralTypes (as already revised in PR #514) to ensure removal of implicit general types that are identical to explicit general types.

Resolutions of the following issues are implemented in this PR:

• SYSML2-79 View::viewpointSatisfactions should subset viewpointChecks and checkedConstraints
• SYSML2-83 Narrow down return types of SpatialItem::PositionOf and ::CurrentPositionOf
• SYSML2-102 Semantic constraint for target of AssignmentActionUsage is missing
• SYSML2-219 Action::decisionTransitions should subset Action::transitions
• SYSML2-305 Message and flow connection ends should be occurrence usages [fully implemented]
• SYSML2-490 Actions::acceptSubactions and sendSubactions should subset acceptActions and sendActions

Resolutions of the following issues were previously implemented in PRs #487, #504 and #506:

• SYSML2-295 Causation end features need to redefine source and target
• SYSML2-305 Message and flow connection ends should be occurrence usages [partially implemented]
• SYSML2-491 KerML constraint requires updates to Systems Library Models
• SYSML2-492 KerML constraint requires updates to Domain Library Models
• SYSML2-497 validateTriggerInvocationExpressionAfterArgument constraint is too strong
• SYSML2-498 validateTriggerInvocationExpressionWhenArgument constraint is wrong
• SYSML2-500 The derivation of AssignmentActionUsage::referent is wrong

Resolutions of the following issues did not require any update to the library model files:

• SYSML2-378 Sample::calculation has an incorrect type
• SYSML2-393 Documentation of Time::defaultClock is missing
• SYSML2-488 Constraints missing to enforce variations being abstract

An additional change was made later in PR #522 consistent with the preliminary resolution to the following issue:

• SYSML2-182 Universal features can have many values

[conradbock]

• action sendSubactions: SendAction[0..*] :> subactions, sendActions action acceptSubactions: AcceptAction[0..*] :> subactions, acceptActions For another issue, presumably these should be intersections, if SysML supports it.

• Connections::FlowConnection, end indenting off.

[seidewitz]

@conradbock

• The SysML notation does not support intersection. And it is not actually necessary here, since these features are not modeled as "sufficient" but, instead, rely on semantic constraints determine their subsetting (like many other features in the semantic libraries).
• The indentation is not consequential. However, I have updated Connections.sysml to consistently use spaces (rather than tabs) for indentation.

[conradbock]

but, instead, rely on semantic constraints determine their subsetting

Do these require all M0 values of subactions that are also values of sendActions will be values of sendSubactions, and vice versa?

And it is not actually necessary here, since these features are not modeled as "sufficient"

Sufficiency doesn't require declaring intersection. It can be used instead of intersection, it's just less well-defined.

the SysML notation does not support intersection.

I figured. :)

[seidewitz]

Do these require all M0 values of subactions that are also values of sendActions will be values of sendSubactions, and vice versa?

They require that all M1 send or accept action usages that are subactions of another action definition or usage subset sendSubactions or acceptSubactions. This may actually miss instances of SendAction and AcceptAction that are not declared using syntactic send or accept actions usages, but that is a general issue with the semantics constraints currently being based on syntactic structure, which can be addressed separately in the future.