fix(observability): re-enable OTLP export for Elixir

Set OTEL_EXPORTER_OTLP_ENDPOINT on the prop deployment to the cluster
Collector so Phoenix/Bandit/Oban spans export and link up with
downstream prop-grid-rs / prop-hrrr-point-rs traces in Tempo. Without
this the Elixir app fell back to traces_exporter :none and the Rust
worker spans had no parent, leaving the trace graph disconnected.

The stale comment referenced opentelemetry_ecto as the reason export
was off, but that package was never added to mix.exs (only
opentelemetry_phoenix / _bandit / _oban are listed), so the
span-flood concern no longer applies.

k8s/deployment.yaml

@@ -61,11 +61,14 @@ spec:
               value: "5000"
             - name: HRRR_BASE_URL
               value: "http://skippy.w5isp.com:8080"
-              # OTLP export disabled: opentelemetry_ecto emits one span per
-              # query, and Oban jobs that run thousands of queries produce
-              # traces Tempo can't ingest (8k+ spans, crashes compactor).
-              # Rust workers still export. Re-enable after opentelemetry_ecto
-              # is removed or filtered.
+            # OTLP/gRPC endpoint for the cluster OTel Collector.
+            # service.name / service.namespace are set in
+            # config/runtime.exs (microwaveprop / prop). Matches the
+            # endpoint used by prop-grid-rs / prop-hrrr-point-rs so
+            # Elixir-initiated traces link up with downstream Rust
+            # worker spans in Tempo.
+            - name: OTEL_EXPORTER_OTLP_ENDPOINT
+              value: "http://otel-collector.observability.svc.cluster.local:4317"
           envFrom:
             - secretRef:
                 name: prop-secrets