Release 0.2.1

[raphaelvallat]

This PR bumps the version to 0.2.1 and adds the changelog entry documenting all changes merged to master since v0.2.0.

What's in this release

Bug fixes

• higuchi_fd: off-by-one in the inner loop — summed N_m - 1 differences per sub-series instead of N_m, causing a systematic underestimate of curve length.
• sample_entropy: returns np.nan (was 0) when no template of length m matches within tolerance — entropy is undefined in that case, not zero.
• epsilon = 10e-9 → 1e-9 in utils.py (silent 10× error in the DFA linear regression guard).
• all = [...] → __all__ = [...] in all three source modules — previously leaked np, jit, KDTree, etc. into the public namespace on from antropy import *.

New feature

• hjorth_params(sf=None): optional sampling frequency parameter converts mobility from samples⁻¹ to Hz.

Docs

• Corrected higuchi_fd docstring formula to match the fix above.
• Added clarifying notes to lziv_complexity, num_zerocross, spectral_entropy, detrended_fluctuation, and hjorth_params.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.00%. Comparing base (e6581bb) to head (dec11a8).
⚠️ Report is 2 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff            @@
##            master       #48   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files            4         4           
  Lines          312       312           
  Branches        52        52           
=========================================
  Hits           312       312           

☔ 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.