spring.ai.chat.client.*.text span tags are not captured when using messages()

[asw12]

Bug description
When ChatClientInputContentObservationFilter is enabled, system text and user text is captured and added as high cardinality key values (spring.ai.chat.client.system.text and spring.ai.chat.client.user.text). But as this filter does so by capturing

• context.getRequest().getSystemText()
• context.getRequest().getUserText()

nothing ends up getting added to the observation when the code is using messages().

Environment
spring-ai 1.0.0-M6
properties="spring.ai.chat.client.observations.include-input=true"

Steps to reproduce
Enable micrometer-observations and set the include-input option to true, as above.

chatClient.prompt()
	.messages(
			new SystemMessage("System Text"),
			new UserMessage("My question"),
			new AssistantMessage("Your answer")
	)
	.call().content();

Expected behavior
I would expect messages() to be split up and treated more similarly to the individual systemText() and userText() calls, where messages() still produces the span tags by concatenating all the systemText and userText together.

This is similar to #873 #2339, where there is a violation of principle of least surprise in that I was not expecting systemText() and userText() to work so differently from messages().

[ThomasVitale]

@asw12 Thanks for raising this issue. It's been addressed in the current snapshot and will be part of the upcoming M8 release (and documented in the upgrade notes). The entire list of messages in the prompt in now included in the span when the feature is enabled.

Relevant bits from the upgrade notes:

The content of a prompt as specified via a ChatClient API was included optionally in the spring.ai.client observation, broken down in a few attributes: spring.ai.chat.client.user.text, spring.ai.chat.client.user.params, spring.ai.chat.client.system.text, spring.ai.chat.client.system.params. All those attributes are now deprecated, replaced by a single gen_ai.prompt attribute that contains all the messages in the prompt, solving the problem affecting the deprecated attributes where part of the prompt was not included in the observation, and aligning with the observations used in the ChatModel API. This new attribute can be enabled via the spring.ai.chat.observations.include-prompt configuration property, whereas the previous spring.ai.chat.observations.include-input configuration property is deprecated.