Added metrics for fission mqtrigger and optimizations in trigger subscriptions (#2399)

- Use mqtrigger watch instead of polling mqtrigger every 5 seconds
- Added metrics to monitor no of subscriptions, and no of messages per subscription
- Add standard go metrics exported by prometheus
- Enable prometheus discovery for mqtrigger pod
- Optimized mqtrigger manager cache
- Add unit tests for mqtrigger cache

Co-authored-by: Sanket Sudake <sanketsudake@gmail.com>

cmd/fission-bundle/mqtrigger/mqtrigger.go

@@ -17,6 +17,7 @@ limitations under the License.
 package mqtrigger
 
 import (
+	"context"
 	"fmt"
 	"os"
 	"path"
@@ -35,7 +36,7 @@ import (
 	_ "github.com/fission/fission/pkg/mqtrigger/messageQueue/nats"
 )
 
-func Start(logger *zap.Logger, routerUrl string) error {
+func Start(ctx context.Context, logger *zap.Logger, routerUrl string) error {
 	fissionClient, _, _, _, err := crd.MakeFissionClient()
 
 	if err != nil {
@@ -74,9 +75,8 @@ func Start(logger *zap.Logger, routerUrl string) error {
 	if err != nil {
 		logger.Fatal("failed to connect to remote message queue server", zap.Error(err))
 	}
-
-	mqtrigger.MakeMessageQueueTriggerManager(logger, fissionClient, mqType, mq).Run()
-
+	mqtMgr := mqtrigger.MakeMessageQueueTriggerManager(logger, fissionClient, mqType, mq)
+	mqtMgr.Run(ctx)
 	return nil
 }