Add BERT Tokenizer as OpenSearch built-in analyzer

[zhichao-aws]

Description

This PR add bert-base-uncased tokenizer and bert-base-multilingual-uncased tokenizer as OpenSearch built-in analyzer/tokenizer. Users can use them via analyze API without doing any special settings:

POST /_analyze
{
    "tokenizer" : "mbert-uncased",
    "text" : "hello world 你好"
}

We also have a follow up PR in neural-search to make neural-sparse search work with the analyzer.

Related Issues

Resolves #3708

Check List

• New functionality includes testing.
• New functionality has been documented.
• API changes companion pull request created.
• Commits are signed per the DCO using --signoff.
• Public documentation issue/PR created.

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.

[zhichao-aws]

We can merge this PR after 3.0.0-beta1 get released

[mingshl]

would you consider option 3 from the issue? #3708 (comment)

[xinyual]

LGTM!

[mingshl]

@zane-neo @xinyual @zhichao-aws we are already code freeze for any new feature PRs. This PR passed the code freeze deadline, would you work with release team to get an exception? or please revert this PR

[zhichao-aws]

Checked with @peterzhuamazon . We can still push feature code to main, as long as not backport to 3.0 branch

[mingshl]

Checked with @peterzhuamazon . We can still push feature code to main, as long as not backport to 3.0 branch

thanks for double checking.