Splicing of default getters should not special case constructors with implicits

[hamzaremmal]

Closes #22061

[hamzaremmal]

Just to keep track of some info, this test is now tests/neg-macros/i12344.scala and it seems that some of the discussions that led to this change are not relevant now (See #12344 (comment) and https://github.com/scala/scala3/pull/14840/files#diff-a602b2ba421d0da906b1d1915ca14c84b336da38e2ab907b695c441c5d58f5b3)

[KacperFKorban]

If you extend the test case to one that also includes the syntactically inserted param lists, you will see that the current fix doesn't work. Test case:

abstract class A()(implicit val x: Int = 3)
class A1 extends A
class A2 extends A()

abstract class B()(using y: Int = 9)
class B1 extends B
class B2 extends B()

abstract class C(implicit val x: Int = 3)
class C1 extends C
class C2 extends C()

abstract class D(using y: Int = 9)
class D1 extends D
class D2 extends D()

[KacperFKorban]

I would probably even extend it to:

abstract class A()(implicit val x: Int = 3)
class A1 extends A
class A2 extends A()

abstract class B()(using y: Int = 9)
class B1 extends B
class B2 extends B()

abstract class C(implicit val x: Int = 3)
class C1 extends C
class C2 extends C()

abstract class D(using y: Int = 9)
class D1 extends D
class D2 extends D()

abstract class E()(implicit val x: Int = 3, val y: Int = 4)
class E1 extends E
class E2 extends E()

abstract class F()(using y: Int = 9, x: Int = 8)
class F1 extends F
class F2 extends F()

abstract class G(implicit val x: Int = 3, val y: Int = 4)
class G1 extends G
class G2 extends G()

abstract class H(using y: Int = 9, x: Int = 8)
class H1 extends H
class H2 extends H()

[KacperFKorban]

I think that we should rather add another condition that checks if meth actually requires the empty argument list i.e.

&& !meth.tpe.widen.isNullaryMethod