What happened?

There is a leaky go routine in Jaeger fswatcher.go library. In fswatcher.go, the new go routine is called whenever New Function is created but that go routine was never terminated.

Steps to reproduce

This only when the new Cassandra Cluster is created directly with no option to close the fswatcher.go goroutine.

Expected behavior

Cassandra Cluster should have a public Close method that allows user to Close the leaky goroutine.

Relevant log output

`[Goroutine 122 in state select, with github.com/jaegertracing/jaeger/pkg/fswatcher.(*FSWatcher).watch on top of the stack:
goroutine 122 [select]:
github.com/jaegertracing/jaeger/pkg/fswatcher.(*FSWatcher).watch(0xc0005d9d40)
        external/com_github_jaegertracing_jaeger/pkg/fswatcher/fswatcher.go:112 +0xc5
created by github.com/jaegertracing/jaeger/pkg/fswatcher.New
        external/com_github_jaegertracing_jaeger/pkg/fswatcher/fswatcher.go:77 +0x125`

Screenshot

No response

Additional context

No response

Jaeger backend version

v1.46.0

SDK

No response

Pipeline

No response

Stogage backend

No response

Operating system

No response

Deployment model

No response

Deployment configs

No response