feat(oss): replace custom resource with l1 cfnindex

[krokoko]

Fixes #1054

• Remove custom resource and Docker dependency for OSS Index
• Use instead the newly released OSS CfnIndex L1 construct

Current status:

• Need to finish implementation (add the fromVectorIndexAttributes and fromCfnIndex methods)
• Review interfaces and include helper methods if needed
• Finish documentation
• Test !

Blockers:
The newly released L1 index prevents feature parity with existing implementation:

• It seems the field data_type (see documentation here: https://opensearch.org/docs/latest/tutorials/vector-search/vector-operations/semantic-search-byte-vectors/) is missing, which provides a way to specify if the embeddings will be stored as binary or float when using with Bedrock KB. We are currently doing this operation using a custom resource but I don't see the needed field exposed through the PropertyMapping field.
• It seems analyzers are not exposed at the moment as well through the IndexSettings. We also configure it through a custom resource for now

---

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of the project license.

[github-actions]

This pull request is now marked as stale because it hasn't seen activity for a while. Add a comment or it will be closed soon. If you wish to exclude this issue from being marked as stale, add the "backlog" label.

[github-actions]

Closing this pull request as it hasn't seen activity for a while. Please add a comment @mentioning a maintainer to reopen. If you wish to exclude this issue from being marked as stale, add the "backlog" label.

[github-actions]

Closing this pull request as it hasn't seen activity for a while. Please add a comment @mentioning a maintainer to reopen. If you wish to exclude this issue from being marked as stale, add the "backlog" label.

[github-actions]

Closing this pull request as it hasn't seen activity for a while. Please add a comment @mentioning a maintainer to reopen. If you wish to exclude this issue from being marked as stale, add the "backlog" label.

[anthony-nhs]

I know this is still in draft, but I have tried using this and one thing I have found is that cloudformation reports the index has been created, but it is not ready for use yet.
This can sometimes lead to a situation where if you try and use the index (in the same stack) in a knowledge base construct, creating the knowledge base fails because the index is not ready yet.