[meshcat] Simplify JointSliders now that all positions have joints

[jwnimmer-tri]

The plant guarantees that all positions are covered by exactly one joint. We no longer need to use a map from position to slider, or deal with non-slider default values.

Towards #22496.

+@RussTedrake for feature review, please.

---

This change is 

[RussTedrake]

Reviewed 2 of 2 files at r1, all commit messages.
Reviewable status: 1 unresolved discussion, needs at least two assigned reviewers

---

multibody/meshcat/joint_sliders.cc line 109 at r1 (raw file):

      Broadcast("lower_limit", -10.0, nq, lower_limit);
  const VectorXd upper_broadcast =
      Broadcast("upper_limit", 10.0, nq, upper_limit);

btw -- now that floating joints are included, using [-1,1] for quaternion values would be dramatically more sensible. i think that's important enough to deserve a special case here. (I actually think that this should be upstreamed as joint limits to MbP, but MbP has asserts that say joint limits are not allowed except for single dof mobilizers).

[jwnimmer-tri]

Reviewable status: needs at least two assigned reviewers

---

multibody/meshcat/joint_sliders.cc line 109 at r1 (raw file):

Previously, RussTedrake (Russ Tedrake) wrote…

btw -- now that floating joints are included, using [-1,1] for quaternion values would be dramatically more sensible. i think that's important enough to deserve a special case here. (I actually think that this should be upstreamed as joint limits to MbP, but MbP has asserts that say joint limits are not allowed except for single dof mobilizers).

Excellent point. The position limit should happen automatically in MbP instead hand-rolling it in many dozens of downstream users of MbP, so for now I've just added a TODO comment (with your name as the person to explain it to others).

[RussTedrake]

Reviewed 1 of 1 files at r2, all commit messages.
Reviewable status: needs at least two assigned reviewers

[SeanCurtis-TRI]

+@SeanCurtis-TRI for platform.

[SeanCurtis-TRI]

Reviewed 1 of 2 files at r1, 1 of 1 files at r2, all commit messages.
Reviewable status: 1 unresolved discussion

---

multibody/meshcat/joint_sliders.cc line 104 at r2 (raw file):

        initial_value->size()));
  }
  const VectorXd nominal_positions =

btw: This could be const reference, thus negating a copy in the case where initial values were provided. But it's initialization, so, meh.

[jwnimmer-tri]

I found a couple more stale comments. PTAL.

[RussTedrake]

Reviewed 1 of 1 files at r3, all commit messages.
Reviewable status: complete! all discussions resolved, LGTM from assignees SeanCurtis-TRI(platform),RussTedrake(platform)