Fix Sentry config to use Sentry.init API

The Sentry setup used a config.sentry.* accessor that sentry-rails has
never provided, so booting with SENTRY_DSN set raised NoMethodError
during environment load (e.g. db:prepare). The code only ran once a DSN
was configured, which is why it surfaced in production now.

Rewrites config/initializers/sentry.rb to call Sentry.init, the actual
sentry-ruby API, and removes the duplicate broken block from
production.rb. Verified production boots with SENTRY_DSN set
(Sentry.initialized? == true) and that the no-DSN path still early-returns.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

config/environments/production.rb

@@ -157,17 +157,5 @@ Rails.application.configure do
   # Skip DNS rebinding protection for the default health check endpoint.
   config.host_authorization = {exclude: ->(request) { request.path == "/up" }}
 
-  # Sentry configuration for production
-  # Only enabled if SENTRY_DSN environment variable is set
-  if ENV["SENTRY_DSN"].present?
-    config.sentry.enabled = true
-
-    # Performance monitoring: sample 20% of transactions for traces
-    # Adjust based on your traffic volume and Sentry plan limits
-    config.sentry.traces_sample_rate = ENV.fetch("SENTRY_TRACES_SAMPLE_RATE", 0.2).to_f
-
-    # Continuous profiling: disabled by default in production due to cost
-    # Enable temporarily for performance investigations if needed
-    config.sentry.profiles_sample_rate = ENV.fetch("SENTRY_PROFILES_SAMPLE_RATE", 0.0).to_f
-  end
+  # Sentry is configured in config/initializers/sentry.rb, gated on SENTRY_DSN.
 end