Varima and AutoVARIMA models for forecasting

[charavelg]

Add VARIMA and AutoVARIMA models for forecasting.

Supports naive seasonality.
Supports variable length TS.

VARIMA uses scipy nelder-mead optimization when q>0
AutoVARIMA uses statsmodels kpss tests for stationarity

Based on https://otexts.com/fpp3/

TODO:

• Improve docstrings
• Add doc example
• Add models into the variable length section in docs

[codecov]

Codecov Report

❌ Patch coverage is 99.75610% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 94.15%. Comparing base (13bf0ea) to head (1fbeccb).
⚠️ Report is 4 commits behind head on main.

Files with missing lines	Patch %	Lines
tslearn/forecasting/_arima.py	99.66%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #665      +/-   ##
==========================================
+ Coverage   93.70%   94.15%   +0.45%     
==========================================
  Files          73       77       +4     
  Lines        6986     7493     +507     
==========================================
+ Hits         6546     7055     +509     
+ Misses        440      438       -2     

☔ 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.