Add langfuse instrumentation

[jcheng5]

The following environment variables must be added to .env:

LANGFUSE_SECRET_KEY
LANGFUSE_PUBLIC_KEY
LANGFUSE_HOST

If these are not available, the app will still work, you just won't get any instrumentation.

See https://langfuse.com/integrations/model-providers/anthropic

[Copilot]

Pull Request Overview

This PR adds Langfuse instrumentation to monitor and trace API calls in the application. The instrumentation is configured to be optional - if the required environment variables are not set, the app will continue to function without instrumentation.

• Adds Langfuse dependencies and OpenTelemetry instrumentation for Anthropic API calls
• Integrates Langfuse client initialization with authentication verification
• Sets up automatic instrumentation for Anthropic API calls using OpenTelemetry

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
shinyapp/requirements.txt	Adds instrumentation dependencies: openai, langfuse, and opentelemetry-instrumentation-anthropic
shinyapp/app.py	Initializes Langfuse client, adds authentication check, and enables Anthropic API instrumentation

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[Copilot]

The error message for authentication failure is not helpful for debugging. Consider providing more specific guidance about checking the LANGFUSE_SECRET_KEY, LANGFUSE_PUBLIC_KEY, and LANGFUSE_HOST environment variables mentioned in the PR description.

Suggested change

	print("Authentication failed. Please check your credentials and host.")
	print(
	"Authentication failed. Please check your LANGFUSE_SECRET_KEY, LANGFUSE_PUBLIC_KEY, and LANGFUSE_HOST environment variables."
	)

[Copilot]

The Anthropic instrumentation is enabled unconditionally, even when Langfuse authentication fails. According to the PR description, instrumentation should be optional when environment variables are not available. Consider wrapping the instrumentation in the authentication check or adding proper error handling.

Suggested change

	else:
	print("Authentication failed. Please check your credentials and host.")
	AnthropicInstrumentor().instrument()
	AnthropicInstrumentor().instrument()
	else:
	print("Authentication failed. Please check your credentials and host.")