Fix Cholesky structural_shock_matrix dim/coord ordering

[niall-stack]

Summary\nFixes the Cholesky identification output to use the same dim/coord ordering as the sign-restriction path (

esponse, shock).\n\n### Changes\n- In Cholesky.identify, changed structural_shock_matrix dims from:\n - (chain, draw, shock, response)\n to:\n - (chain, draw, response, shock)\n- Updated coord mapping accordingly.\n- Added a regression assertion in ests/test_identification.py to lock the expected dim + coord order.\n\n## Why\nIssue #26 reports that Cholesky xarray labels are transposed relative to SignRestriction, which can cause downstream confusion/misalignment.\n\n## Validation\nAdded targeted test assertions for dims and coords.\n\nNote: I couldn’t execute pytest in this environment because no Python runtime is installed on host (py -3.11 reports no suitable runtime).

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 88.9%. Comparing base (cfe95cc) to head (91a2d62).
⚠️ Report is 1 commits behind head on main.

❌ Your project status has failed because the head coverage (88.9%) is below the target coverage (90.0%). You can increase the head coverage or adjust the target coverage.

Additional details and impacted files

@@          Coverage Diff          @@
##            main     #29   +/-   ##
=====================================
  Coverage   88.9%   88.9%           
=====================================
  Files         16      16           
  Lines        651     651           
  Branches      83      83           
=====================================
  Hits         579     579           
  Misses        53      53           
  Partials      19      19           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[thomaspinder]

Thanks for opening this PR. It looks like the PR branch is stale, so I think you'll just need to rebase the PR. Once this is done, I think we can merge.