BUG: Ensure std_resid returns named pandas Series for numpy array inputs

[samanvirajput]

Summary

Fixes type inconsistency in ARCHModelResult.std_resid when model input is a numpy array.

Problem

When arch_model() receives a numpy ndarray, std_resid returns a raw numpy array
with no .name attribute instead of a named pandas Series.

Reproducer:

import numpy as np
from arch import arch_model
res = arch_model(np.random.randn(500), vol='GARCH').fit(disp='off')
print(type(res.std_resid))  # <class 'numpy.ndarray'> — should be pd.Series

Fix

Added else branch to wrap numpy output in a named pd.Series with the model index.

Tests

Added 3 tests to arch/tests/univariate/test_base.py covering numpy input,
pandas input (backwards compat), and index preservation.

Motivation

Discovered while computing GARCH residual diagnostics for fat-tail analysis
of Indian equity markets: https://doi.org/10.5281/zenodo.20573035

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.54%. Comparing base (d61843e) to head (efb475d).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #844   +/-   ##
=======================================
  Coverage   99.54%   99.54%           
=======================================
  Files          78       78           
  Lines       15817    15846   +29     
  Branches     1294     1294           
=======================================
+ Hits        15745    15774   +29     
  Misses         38       38           
  Partials       34       34           

Flag	Coverage Δ
adder	99.50% <100.00%> (+<0.01%)	⬆️
subtractor	99.50% <100.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ 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.