[FIX] ASR: return an estimate when geometric_median does not converge

[sappelhoff]

Summary

When the Vardi–Zhang iteration in geometric_median reached max_iter without
meeting the tolerance, the while ... else branch only printed a message and
the function fell through to an implicit return None, which breaks callers
(asr_calibrate / asr_process reshape the result). Return the last iterate
y instead, which is the best estimate available.

The message wording is also fixed ("could converge" → "could not converge").

Correctness

• The converged path is unchanged.
• On non-convergence the function now returns a correctly-shaped array instead
  of None (verified with a forced single-iteration run).

[codecov]

Codecov Report

❌ Patch coverage is 0% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 83.10%. Comparing base (1606601) to head (df34644).
⚠️ Report is 1 commits behind head on master.

Files with missing lines	Patch %	Lines
meegkit/utils/asr.py	0.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #107      +/-   ##
==========================================
- Coverage   83.13%   83.10%   -0.03%     
==========================================
  Files          25       25              
  Lines        2817     2818       +1     
==========================================
  Hits         2342     2342              
- Misses        475      476       +1     

☔ View full report in Codecov by Harness.
📢 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.