[Feat] Refactor and finalize Run service

[machichima]

Description

There are few tasks left in run service:

1. Some function pass protobuf request directly to the repository. This violates separation of concerns and creates a dependency from the repository layer to the API layer. We should refactor this to follow how we did in task_service

flyte/runs/service/run_service.go

Line 53 in 5c20324

run, err := s.repo.ActionRepo().CreateRun(ctx, req.Msg)

flyte/runs/service/task_service.go

Line 66 in 5c20324

err = s.db.TaskRepo().CreateTask(ctx, taskModel)

2. There's some TODOs left that we need to build the response from the run model

flyte/runs/service/run_service.go

Lines 163 to 169 in 5c20324

	// TODO: Build full RunDetails from the run model
	// For now, return a minimal response
	resp := &workflow.GetRunDetailsResponse{
	Details: &workflow.RunDetails{
	// Would populate this from run model
	},
	}

Can also follow what we have in task service

flyte/runs/service/task_service.go

Line 82 in 5c20324

tasks, err := transformers.TaskModelsToTaskDetailsWithoutIdentity(ctx, []*models.Task{model})

[runllm]

It seems likely that the Run service in Flyte currently passes protobuf request objects directly to the repository layer, which is not ideal for separation of concerns. The recommended approach is to refactor the Run service to decouple the API (protobuf) layer from the repository layer, following the pattern used in the Task service, where conversion between API and model types happens in the service layer, not the repository layer. This is supported by the code structure and comments in the referenced files, and is consistent with recent refactors in the Flyte codebase (see PR #6813).

Additionally, there are TODOs in the Run service for building responses from the run model, which should be addressed by implementing response construction similar to the Task service. The Task service provides a clear example of how to map model objects to API responses, and this pattern should be replicated in the Run service for consistency and maintainability (see PR #6813). Would you like more detail on how to implement these refactors step-by-step?

Sources:

• see PR #6813

---

This is an AI-generated response and your feedback is appreciated! Please leave a 👍 if this is helpful and 👎 if it is not.

Mention @RunLLM to ask follow-up questions. See runllm.com for more!

[yuhuan130]

#take