feat(T1.2): implement structured logging and Prometheus metrics

- Add internal/logging package with zap-based structured JSON logging
- Support development (colored) and production (JSON) modes via ENVIRONMENT env var
- Add logging helpers: Info(), Error(), Warn(), Debug(), Fatal()
- Add field helpers: String(), Int(), Int64(), Err()
- Add internal/metrics package with 16 comprehensive Prometheus metrics
- Track workflows: starts, completions, duration by type/status
- Track activities: starts, completions, duration, retries by type
- Track LLM calls: total calls and latency by model
- Track git operations: total and duration by operation type
- Track judge decisions: decisions by type
- Track Temporal errors: connection errors by type
- Track cache efficiency: hits and misses by cache type
- Track tasks in progress: gauge metric by task type
- Metrics exported on /metrics endpoint (Prometheus text format)
- Integrate structured logging in cmd/worker and cmd/starter
- Replace all log.Printf/log.Fatalf with structured logging
- Add /metrics endpoint to health check server
- 8/8 logging tests passing, 13/13 metrics tests passing
- All verification criteria met

Dependencies added:
- go.uber.org/zap v1.28.0 (structured logging)
- github.com/prometheus/client_golang v1.24.1 (metrics export)

Closes T1.2

internal/metrics/metrics_test.go

@@ -0,0 +1,111 @@
+package metrics
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestRecordWorkflowStarted(t *testing.T) {
+	// Should not panic
+	assert.NotPanics(t, func() {
+		RecordWorkflowStarted("TestWorkflow")
+	})
+}
+
+func TestRecordWorkflowCompleted(t *testing.T) {
+	// Should not panic
+	assert.NotPanics(t, func() {
+		RecordWorkflowCompleted("TestWorkflow", "success", 1.5)
+	})
+}
+
+func TestRecordActivityStarted(t *testing.T) {
+	// Should not panic
+	assert.NotPanics(t, func() {
+		RecordActivityStarted("TestActivity")
+	})
+}
+
+func TestRecordActivityCompleted(t *testing.T) {
+	// Should not panic
+	assert.NotPanics(t, func() {
+		RecordActivityCompleted("TestActivity", "success", 0.5)
+	})
+}
+
+func TestRecordActivityRetry(t *testing.T) {
+	// Should not panic
+	assert.NotPanics(t, func() {
+		RecordActivityRetry("TestActivity")
+	})
+}
+
+func TestRecordLLMAPICall(t *testing.T) {
+	// Should not panic
+	assert.NotPanics(t, func() {
+		RecordLLMAPICall("claude-opus", "success", 2.0)
+	})
+}
+
+func TestRecordGitOperation(t *testing.T) {
+	// Should not panic
+	assert.NotPanics(t, func() {
+		RecordGitOperation("clone", "success", 5.0)
+	})
+}
+
+func TestRecordJudgeDecision(t *testing.T) {
+	// Should not panic
+	assert.NotPanics(t, func() {
+		RecordJudgeDecision("approve")
+	})
+}
+
+func TestRecordTemporalConnectionError(t *testing.T) {
+	// Should not panic
+	assert.NotPanics(t, func() {
+		RecordTemporalConnectionError("connection_timeout")
+	})
+}
+
+func TestRecordCacheHit(t *testing.T) {
+	// Should not panic
+	assert.NotPanics(t, func() {
+		RecordCacheHit("llm_response")
+	})
+}
+
+func TestRecordCacheMiss(t *testing.T) {
+	// Should not panic
+	assert.NotPanics(t, func() {
+		RecordCacheMiss("llm_response")
+	})
+}
+
+func TestUpdateTasksInProgress(t *testing.T) {
+	// Should not panic
+	assert.NotPanics(t, func() {
+		UpdateTasksInProgress("T0", 5.0)
+	})
+}
+
+// TestMetricsExist verifies all metrics are registered
+func TestMetricsExist(t *testing.T) {
+	assert.NotNil(t, WorkflowExecutionsStarted)
+	assert.NotNil(t, WorkflowExecutionsCompleted)
+	assert.NotNil(t, WorkflowDuration)
+	assert.NotNil(t, ActivityExecutionsStarted)
+	assert.NotNil(t, ActivityExecutionsCompleted)
+	assert.NotNil(t, ActivityDuration)
+	assert.NotNil(t, ActivityRetries)
+	assert.NotNil(t, LLMAPICallsTotal)
+	assert.NotNil(t, LLMAPILatency)
+	assert.NotNil(t, GitOperationsTotal)
+	assert.NotNil(t, GitOperationsDuration)
+	assert.NotNil(t, TasksInProgress)
+	assert.NotNil(t, JudgeDecisionsTotal)
+	assert.NotNil(t, TemporalConnectionErrors)
+	assert.NotNil(t, CacheHits)
+	assert.NotNil(t, CacheMisses)
+}