Pull request: #2715

Pull request: #2717

Pull request: #2716

Pull request: #2718

Pull request: #2720

Add a new CLI option `--meta-level` accepting an `Int`. Default is `0`
and means the root project, `1` is the parent meta-build, if defined, or
the built-in bootstrap module, and so on.

**Example: Find version updates in the meta build**

Here is some example output (applied to the mill repo):

``` 
$ mill --meta-level 1 mill.scalalib.Dependency/showUpdates
[1657/1657] dev.run 
Found 3 dependency update for 
  net.sourceforge.htmlcleaner:htmlcleaner : 2.25 -> 2.26 -> 2.27 -> 2.28 -> 2.29
  com.lihaoyi:mill-contrib-buildinfo_2.13 : 0.11.2-6-261437 -> 0.11.2
  com.github.lolgab:mill-mima_mill0.11_2.13 : 0.0.23 -> 0.0.24
```

**Meta information about the build**

I also added a new external module `mill.runner.MillBuild` to get some
meta-information about the project, for now, the meta-module count or
frame count.

Here on a project with one meta-build:

```
$ mill show mill.runner.MillBuild/levelCount
3
```

Pull request: #2719

Pull request: #2722

Pull request: #2726

Pull request: #2727

Pull request: #2728

This PR adds an addition suggestion to the `Try ... to see what's
available` error message based on the task names present in the entire
build.

Before: 

```
$ ./mill -i dev.run example/scalabuilds/10-scala-realistic -i Compilee
Cannot resolve Compilee. Try `mill resolve _` to see what's available.
```

After: 
```
$ ./mill -i dev.run example/scalabuilds/10-scala-realistic -i Compilee
Cannot resolve Compilee. Try `mill resolve _` or `mill resolve __.compile` to see what's available.
```

This should help in the common case where someone runs a task with the
correct name but in the wrong module. We do a fuzzy match, so even if
there's a small typo we still are able to give a suggestion. This is
complementary to the existing help message, which is useful when the
user runs a task with the wrong name but in the correct module

The way this works is that we capture all the non-command task names in
the `Discover` macro, and when a user runs a task that fails to resolve,
we look at the last segment of the query and run a fuzzy search against
the set. If a fuzzy match is found, we suggest to the user to run `mill
resolve __.foo` so they have a way to see what the fully-qualified name
of the task they were trying to run, and run it. Given the typical build
would have O(10s) to O(100s) of modules at most, we can expect even the
most verbose `mill resolve __.foo` outputs to be small enough that the
user can browse through and find what they want

Fixes #2732

We introduce a new `localCompileClasspath`, which differs from
`compileClasspath` in that it leaves out the
`transitiveCompileClasspath` from upstream modules and `resolvedIvyDeps`
from third party dependencies.

Also tried to do some cleanup of `JavaModule`, refactoring out a
`transitiveModuleCompileModuleDeps` helper to DRY things up,
re-arranging things so all the `fooModuleDeps` helpers are all together.

Tested via a new unit test
`mill.scalalib.HelloWorldTests.multiModuleClasspaths`, which fails on
master and passes on this PR. This test case asserts the overall "shape"
of the classpaths, which should help rule out an entire class of
mis-configuration w.r.t. how the classpaths are wired up

In Scala 3 modules, that also define a platform, the scala3-library is
not correctly resolved.
The current implementation exists merely as a convenience for JS and
Native modules. `ScalaNativeModule` already handles a correct scala
library by overriding `scalaLibraryIvyDeps`. This pull request overrides
the same in `ScalaJSModule` and removed the faulty platfrom dependency
from `ScalaModule`.

E.g. imagine, someone tries to build a Mill plugin with Scala 3 (which
is currently not supported) and sets the platform to `_mill0.11`, then
you would run into the following issue:

```scala
object root extends ScalaModule {
  override def scalaVersion = "3.3.0"
  override def platformSuffix = "_mill0.11"
  // ..
}
```

```
>> mill ivyDepsTree
1 targets failed
resolvedIvyDeps
Resolution failed for 1 modules:
--------------------------------------------
  org.scala-lang:scala3-library_mill0.11_3:3.3.0
	not found: /Users/me/ivy2/local/org.scala-lang/scala3-library_mill0.11_3/3.3.0/ivys/ivy.xml
	not found: https://repo1.maven.org/maven2/org/scala-lang/scala3-library_mill0.11_3/3.3.0/scala3-library_mill0.11_3-3.3.0.pom
```

Pull request: #2739

Pull request: #2740

Pull request: #2738

Tries to fix #2704 by ensuring
the system streams are properly inherited when running `.console` or
`.repl` with `-i`.


Tested via

```bash
> ./mill show dev.launcher

> cd example/basic/1-simple-scala

> /Users/lihaoyi/Github/mill/out/dev/launcher.dest/run  -i console
```

Notably this cannot be tested via `dev.run`, because we would need to
apply the same `SystemStreams.withStreams(SystemStreams.original)`
wrapper to `dev.run`, which would first require re-bootstrapping Mill on
top of this PR