Bug 2036474 - Updated s2s telemetry

[fmasalha]

No description provided.

[github-actions]

View this pull request in Lando to land it once approved.

[segunfamisa]

I think that if we have a future where we have multiple model id's, or a dynamic model id, this would break.

If I remember correctly, we have the model id as part of the request.......so I don't know if it is "static". I believe it could be dynamic, so far...based on the design we have.

[segunfamisa]

I believe that we have a fixed value we define here -

firefox/mobile/android/android-components/components/lib/llm-mlpa/src/main/java/mozilla/components/lib/llm/mlpa/MlpaLlm.kt

Lines 26 to 34 in f63224e

	internal val Prompt.asRequest
	get() = Request(
	model = ModelID.mozSummarization,
	messages = buildList {
	systemPrompt?.let { add(Message.system(it)) }
	add(Message.user(userPrompt))
	},
	stream = true,
	)

Is there a way we could pass the model from the llmProvider to the Llm Request? So that we are certain that the correct model owned by the provider is the same one used for the prompt request

[boek]

Since this is a telemetry patch let's move it to phabricator