Fix model checker errors

[lhartl]

To be able to use the model checker, the existing model checker errors and warnings should be fixed where possible.

[heikob2]

Hi, i have a look at the model-checker errors in tests.in.expr.os:

• maybe a bug in the TS: shouldn't be type number also be of type any testsuite AlgebraicLanguage::eval()
• isn't int and real deprecated?
• IListOneArgOp gives TS-error, if arg is a list and number of entries do not match (with "left" operand).
• maybe there is a bug in IMetaFunction::isActualFunctionValid(), found in helloQuery (groupedAndSummed and groupedAndCounted ) visible in inspector; quickfix has no effect.

See branch fix/modelchecker-testInExprOs

[wsafonov]

maybe a bug in the TS: shouldn't be type number also be of type any testsuite AlgebraicLanguage::eval()

To me any seems to be a candidate to be a supertype of everything, but right now it's only supertype of join types. Maybe @markusvoelter can give some hints why this has been done this way.

But we clearly have here an issue with deriving a common type for match expressions - in my case it gets a boolean type.

isn't int and real deprecated?

No, they are just simple way of expressing integer and floating-point number types.

IListOneArgOp gives TS-error, if arg is a list and number of entries do not match (with "left" operand).

Looks like a TS bug with the ListWithAllOp that doesn't checks its arg for a list type, instead just generally expecting its type to be a subtype of the base list type which is not correct (due to a possible list size mismatch as you noted)