Prepare opensearch-py for 3.0.0 release

[saimedhi]

Description

• Updated API Generator within this repo as needed for the latest opensearch api specification structure changes

• Enforced mandatory keyword-only arguments for calling auto-generated OpenSearch-py APIs

• Removing deprecated APIS and corresponding tests

• Made Version bump changes for 3.0.0. release.

• Merged changes from latest generated api code.

• Cleaned up guides and samples that are using positional arguments instead of keyword arguments.

• Updated github workflows to run tests against latest opensearch versions.

• Fixed around 120 failing tests both unit and integration tests combined.

• Relevant PRs in other repos:

  • Add request body to search_shards api opensearch-api-specification#903
  • fix indices.simulate_index_template test spec OpenSearch#18399

• Link checker failing expected as a part of version bump to 3.0.0.

Issues Resolved

Closes #903
Closes #904
Closes #898
Closes #867
Closes #413

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.

[codecov]

Codecov Report

Attention: Patch coverage is 48.74608% with 327 lines in your changes missing coverage. Please review.

Project coverage is 68.96%. Comparing base (ba715b9) to head (ddba686).
Report is 84 commits behind head on main.

Files with missing lines	Patch %	Lines
opensearchpy/_async/plugins/ml.py	43.68%	116 Missing ⚠️
opensearchpy/plugins/ml.py	43.68%	116 Missing ⚠️
opensearchpy/_async/plugins/sm.py	44.18%	24 Missing ⚠️
opensearchpy/_async/plugins/geospatial.py	48.57%	18 Missing ⚠️
opensearchpy/plugins/geospatial.py	48.57%	18 Missing ⚠️
opensearchpy/_async/plugins/ltr.py	58.33%	15 Missing ⚠️
opensearchpy/plugins/ltr.py	58.33%	15 Missing ⚠️
opensearchpy/_async/helpers/actions.py	0.00%	1 Missing ⚠️
opensearchpy/_async/helpers/document.py	0.00%	1 Missing ⚠️
opensearchpy/_async/plugins/neural.py	83.33%	1 Missing ⚠️
... and 2 more

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #907      +/-   ##
==========================================
- Coverage   71.95%   68.96%   -2.99%     
==========================================
  Files          91      131      +40     
  Lines        8001     9900    +1899     
==========================================
+ Hits         5757     6828    +1071     
- Misses       2244     3072     +828     

☔ 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.

[saimedhi]

Failing tests are due to actions/runner-images#12273 and not related to this PR

[saimedhi]

Tracking ci tests hang indefinitely issue here actions/runner-images#12322

[dblock]

We can't merge anything until CI is green. If code coverage needs to be disabled or something else that doesn't make sense to run, go for it.

[saimedhi]

Tracking ci tests hang indefinitely issue here actions/runner-images#12322

Previous issue for integ tests with security is fixed. But few other tests previously passing are now getting hang. Issue is occuring when unit tests are running on windows. Will report back to actions/runner-images.

CI is now green and passing

[roharvey]

@saimedhi can you add detail about this line?

Enforced mandatory keyword-only arguments for calling auto-generated OpenSearch-py APIs

I wonder because many of the previous functions had some non-default arguments (e.g. index in indices.create()), so the first positional argument was required in any case. Likewise with indices.delete() - what else are you going to put in the first argument other than an index name? Was this a preferential coding style change or was there something external which prompted the keyword-only args? It caused quite a number of regressions when I tested the upgrade.

[Pigueiras]

Making breaking changes is OK, but squashing a breaking change into a 'Prepare 3.0.0 release' commit and not mentioning it in the release notes is evil 🤣

[dblock]

@Pigueiras I am late to the game here, but I 100% agree.