Update sparse text embeddings API route for Inference Service

[demjened]

Per discussion with the Search Inference team, we're modifying the API path of ELSER embeddings in Elastic Inference Service: /api/v1/sparse-text-embedding/ELSERv2 -> /api/v1/sparse-text-embedding*s*/ELSERv2.

[elasticsearchmachine]

Hi @demjened, I've created a changelog YAML for you.

[elasticsearchmachine]

Pinging @elastic/search-inference-team (Team:Search - Inference)

[elasticsearchmachine]

Pinging @elastic/search-eng (Team:SearchOrg)

[maxjakob]

Hold on a minute, are we sure we want the task type to be different to the Inference API? It is called sparse_embedding there.

[demjened]

@maxjakob IMO there is no need for the downstream Inference Service path to match the path of the Inference API - that's an implementation detail in the model class. Also, we're adding /api/v1/chat/completions in EIS which is plural on purpose so that it mirrors OpenAI's spec. It's also more RESTful (resources are plural).

So in EIS we can adopt either naming pattern as long as we're consistent.

[elasticsearchmachine]

💔 Backport failed

Status	Branch	Result
❌	8.16	Commit could not be cherrypicked due to conflicts
❌	8.17	Commit could not be cherrypicked due to conflicts
❌	8.x	Commit could not be cherrypicked due to conflicts

You can use sqren/backport to manually backport by running backport --upstream elastic/elasticsearch --pr 118025