[Semantic Text UI] Handle the case when the model is not yet downloaded

[saikatsarkar056]

When the trained model is not yet downloaded, it can't be deployed. This PR has covered the following:

• Download the model if it does not exist
• Tests to support this change

How to test the changes locally

• Download the elasticsearch changes from GitHub branch
• Run the elasticsearch: ./gradlew :run -Drun.license_type=trial
• Download the changes of this PR in local kibana and do the following steps
  • Set isSemanticTextEnabled = true in this location
  • Run yarn start

[saikatsarkar056]

@elastic/kibana-management It would be great if I can get a review for this PR.

[saikatsarkar056]

@elasticmachine merge upstream

[sphilipse]

one nitpick but generally looks good :)

[yuliacech]

Thanks a lot for addressing this use case, @saikatsarkar056!
After testing the changes locally, I'd just want to confirm that the UX with the modal blocking the user from saving the mappings is the best choice. If I understand correctly, the modal will now wait until the model is downloaded and deployed. That could take a really long time. Should we consider starting the download and deployment of the model in the background instead?

[saikatsarkar056]

Thanks a lot for addressing this use case, @saikatsarkar056! After testing the changes locally, I'd just want to confirm that the UX with the modal blocking the user from saving the mappings is the best choice. If I understand correctly, the modal will now wait until the model is downloaded and deployed. That could take a really long time. Should we consider starting the download and deployment of the model in the background instead?

@yuliacech Please notice this code-block. The model download and deployment are occurring in the background. Only the check to determine if a model has already been downloaded is performed asynchronously, which is quite fast.

Furthermore, I've relocated the CreateInferenceEndpoint block to its own function, allowing us to call it after the model has been downloaded and deployed.

The modifications in this pull request indicate that the model download, model deployment, and CreateInferenceEndpoint processes occur one-after-another in the background.

[saikatsarkar056]

@yuliacech Thank you for the feedback. Can you please review this PR once again?

[kibana-ci]

💚 Build Succeeded

• Buildkite Build
• Commit: 55cf7e1

Metrics [docs]

Public APIs missing comments

Total count of every public API that lacks a comment. Target amount is 0. Run node scripts/build_api_docs --plugin [yourplugin] --stats comments for more detailed information.

id	before	after	diff
ml	66	67	+1

Async chunks

Total size of all lazy-loaded chunks that will be downloaded as the user navigates the app

id	before	after	diff
indexManagement	655.2KB	655.8KB	+629.0B

Public APIs missing exports

Total count of every type that is part of your API that should be exported but is not. This will cause broken links in the API documentation system. Target amount is 0. Run node scripts/build_api_docs --plugin [yourplugin] --stats exports for more detailed information.

id	before	after	diff
ml	100	101	+1

Unknown metric groups

API count

id	before	after	diff
ml	153	154	+1

History

• 💚 Build #208528 succeeded d02d315
• 💚 Build #207987 succeeded 88b1064
• 💚 Build #206768 succeeded d27305f

To update your PR or re-run it, just comment with:
@elasticmachine merge upstream

cc @saikatsarkar056

[jgowdyelastic]

ML changes LGTM