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/logging/logger_test.go

@@ -0,0 +1,74 @@
+package logging
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestInitLogger(t *testing.T) {
+	err := InitLogger()
+	assert.NoError(t, err)
+}
+
+func TestGetLogger(t *testing.T) {
+	lg := GetLogger()
+	assert.NotNil(t, lg)
+}
+
+func TestStringField(t *testing.T) {
+	field := String("key", "value")
+	assert.NotNil(t, field)
+	assert.Equal(t, "key", field.Key)
+}
+
+func TestIntField(t *testing.T) {
+	field := Int("counter", 42)
+	assert.NotNil(t, field)
+	assert.Equal(t, "counter", field.Key)
+}
+
+func TestInt64Field(t *testing.T) {
+	field := Int64("bignum", 9223372036854775807)
+	assert.NotNil(t, field)
+	assert.Equal(t, "bignum", field.Key)
+}
+
+func TestErrorField(t *testing.T) {
+	err := assert.AnError
+	field := Err(err)
+	assert.NotNil(t, field)
+	assert.Equal(t, "error", field.Key)
+}
+
+// Note: TestSync is omitted because zap.Sync() may fail on stderr in test environment
+// This is expected behavior and doesn't affect production use
+
+func TestWith(t *testing.T) {
+	InitLogger()
+	lg := With(String("test", "value"))
+	assert.NotNil(t, lg)
+}
+
+// TestLoggingFunctions tests that logging functions don't panic
+func TestLoggingFunctions(t *testing.T) {
+	InitLogger()
+	defer Sync()
+
+	// These should not panic
+	assert.NotPanics(t, func() {
+		Info("test info", String("field", "value"))
+	})
+
+	assert.NotPanics(t, func() {
+		Warn("test warn", String("field", "value"))
+	})
+
+	assert.NotPanics(t, func() {
+		Debug("test debug", String("field", "value"))
+	})
+
+	assert.NotPanics(t, func() {
+		Error("test error", String("field", "value"))
+	})
+}