chore: Switch to new new wildcard [?].

[He-Pin]

Motivation:
Test and migrate code to use ?.

known issue: VirtusLab/scala-cli#2684

Note: this pr is for testing

background: #878

[error] -- Error: /home/runner/work/incubator-pekko/incubator-pekko/actor/src/main/scala/org/apache/pekko/actor/TypedActor.scala:277:38 
[error] 277 |      interfaces: immutable.Seq[Class[_]])
[error]     |                                      ^
[error]     |`_` is deprecated for wildcard arguments of types: use `?` instead
[error]     |This construct can be rewritten automatically under -rewrite -source 3.4-migration.

[He-Pin]

Will need another ignore blame, if this can get merged.

[mdedetrich]

So I am a bit on the fence of this, mainly because its such a massive rewrite that its going to cause quite a bit of chaos especially with the amount of open PR's we have. I would be more amicable to it if we only rewrote scala-3 sources rather than the entire project.

Another disadvantage is that it will make backports to 1.0.x slightly more complicated (I guess we can also apply this migration to the 1.0.x branch to solve this specific issue as its only syntax that doesn't have any other impact).

I am not against it, but there might be better timing to merge such a PR (i.e. when there are very few open PR's). Curious to hear what others think. Would also like to see the response from VirtusLab/scala-cli#2684

[He-Pin]

If I compile it with Scala 3.4, there will be many warning, I want our code prepare for the next LTS early, at least seems not that harm, and in Java we use <?> everyday.

[mdedetrich]

If I compile it with Scala 3.4, there will be many warning, I want our code prepare for the next LTS early, at least seems not that harm, and in Java we use <?> everyday.

Thanks for providing the context, then I would initially suggest lets wait until Pekko 1.1.0 is out and some more PR's are merged, Scala 3.5.x LTS is still a long way off.

[He-Pin]

scala/scala3#19479

@mdedetrich another case why I want pekko code to be more scala 3.4 friendly

[mdedetrich]

@mdedetrich another case why I want pekko code to be more scala 3.4 friendly

This appears to be entirely unrelated and type inferencer/compiler bug, not syntax.