[primitives] Add input_required option to PassThrough

[castor639]

Resolves #21727 .

---

This change is 

[ggould-tri]

Assigning +a:@sherm1 as systems framework area lead; feel free to delegate. Noted as release-notable since at superficial glance I think this is a breaking API change to the primitives.

@ggould-tri made 1 comment.
Reviewable status: LGTM missing from assignee sherm1(platform), needs at least two assigned reviewers (waiting on sherm1).

[sherm1]

@drake-jenkins-bot test this please

[sherm1]

The code looks fine and I don't think it's a breaking change, but I am unclear on the motivation for this change? Do you have a use case for it? Please explain why we should make this change now.

@sherm1 made 1 comment.
Reviewable status: LGTM missing from assignee sherm1(platform), needs at least two assigned reviewers.

[sherm1]

@drake-jenkins-bot retest this please

[jwnimmer-tri]

For context -- this is a fix for #21727.

[sherm1]

OK, thanks Jeremy. @castor639 when a PR addresses some issue, please put the issue number in the PR description, preceded by GitHub magic like "Fixes" or "Resolves" so that when the PR is merged the issue is automatically closed.

[sherm1]

Feature , with one nit
+a:@SeanCurtis-TRI for platform review per rotation (Tuesday)

@sherm1 reviewed 6 files and all commit messages, and made 2 comments.
Reviewable status: 1 unresolved discussion, LGTM missing from assignee SeanCurtis-TRI(platform) (waiting on castor639 and SeanCurtis-TRI).

---

systems/primitives/pass_through.h line 56 at r1 (raw file):

  /// connected will throw. If false (the default), the zero default value is
  /// used instead.
  /// @pydrake_mkdoc_identifier{1args_vector_size}

nit: these "1args" mkdoc names seem wrong for these 2-arg methods. Not clear to me why you need the arg count in the name anyway. Consider just vector_size, value, and abstract_model_value as the mkdoc names.