Locks qpsolvers version to be last known compatible version with PINK IK

[peterd-NV]

Description

Locks the version of qpsolvers to 4.11.0 which is last known version that works with pin_pink.
This fixes Pink IK tests.

Type of change

• Bug fix (non-breaking change which fixes an issue)

Checklist

• I have read and understood the contribution guidelines
• I have run the pre-commit checks with ./isaaclab.sh --format
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• I have updated the changelog and the corresponding version in the extension's config/extension.toml file
• I have added my name to the CONTRIBUTORS.md or my name already exists there

[greptile-apps]

Greptile Summary

This PR pins qpsolvers to 4.11.0 to prevent pin-pink==3.1.0 from breaking due to the removal of the primal_start keyword argument in qpsolvers 4.12.0, which caused PinkIKController to silently return no-op joint positions.

• setup.py: adds qpsolvers==4.11.0 under the same Linux ARM/x86_64 PEP 508 marker block as the other pink_ik dependencies (pin, pin-pink==3.1.0, daqp==0.7.2).
• changelog.d/peterd-lock-qpsolvers-version.rst: new fragment documenting the root cause and the silent IK failure it produced.

Confidence Score: 5/5

Safe to merge; the change is a one-line dependency pin with no logic modifications.

The only change is adding qpsolvers==4.11.0 under the same platform guard already used for the other pink_ik dependencies. It correctly prevents pip from picking up the incompatible 4.12.0 release that caused a silent IK no-op. The changelog fragment is accurate and follows the existing format.

No files require special attention; the changelog entry has a minor style inconsistency (missing Changed section) but does not affect functionality.

Important Files Changed

Filename	Overview
source/isaaclab/setup.py	Adds qpsolvers==4.11.0 as an explicit dependency under the same Linux + ARM/x86_64 PEP 508 marker block used by the other pink_ik dependencies, preventing pip from resolving a newer incompatible version.
source/isaaclab/changelog.d/peterd-lock-qpsolvers-version.rst	New changelog fragment documenting the fix; follows the Fixed section convention but is missing a Changed section to record the added install requirement, unlike the pattern used in comparable entries.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A["pip install isaaclab (Linux ARM/x86)"] --> B["Resolves qpsolvers"]
    B -- "Before PR (unconstrained)" --> C["qpsolvers>=4.12.0 installed"]
    C --> D["pin-pink==3.1.0 calls solve_ik()"]
    D --> E["TypeError: unexpected kwarg 'primal_start'"]
    E --> F["PinkIKController silently returns current joints (IK no-op)"]
    B -- "After PR (qpsolvers==4.11.0)" --> G["qpsolvers 4.11.0 installed"]
    G --> H["pin-pink==3.1.0 calls solve_ik()"]
    H --> I["IK solved correctly"]

Loading

_{Reviews (1): Last reviewed commit: "lock qpsolvers version to be last known ..." | Re-trigger Greptile}

[greptile-apps]

Other entries that add a new install requirement (e.g., antoiner-fix-from-files-windows-filelock.rst) include a Changed section alongside the Fixed section to record the added dependency. Adding one here would keep the changelog consistent.

Suggested change

	* Pinned ``qpsolvers==4.11.0`` to keep Pink IK working. ``qpsolvers`` 4.12.0
	dropped the ``primal_start`` kwarg, causing ``pin-pink==3.1.0`` to raise
	``TypeError`` inside ``solve_ik``;
	:class:`~isaaclab.controllers.pink_ik.PinkIKController` then silently
	fell back to returning the current joints, making IK a no-op.
	* Pinned ``qpsolvers==4.11.0`` to keep Pink IK working. ``qpsolvers`` 4.12.0
	dropped the ``primal_start`` kwarg, causing ``pin-pink==3.1.0`` to raise
	``TypeError`` inside ``solve_ik``;
	:class:`~isaaclab.controllers.pink_ik.PinkIKController` then silently
	fell back to returning the current joints, making IK a no-op.
	Changed
	^^^^^^^
	* Added ``qpsolvers==4.11.0`` to ``isaaclab`` install requirements (Linux only).

[kellyguo11]

Looks like @ooctipus was able to fix the tests in #5556. @peterd-NV do those changes look good to you?