[WIP] add and use tracing API as part of package logging

- make `logging.GetLogger(ctx, Subsys)` the authoritative `logger.Logger` factory function
    - the context carries a linked list of injected fields which
      `logging.GetLogger` adds to the logger it returns
- introduce the concept of tasks and spans, also tracked as linked list within ctx
    - [ ] TODO automatic logging of span begins and ends, with a unique
      ID stack that makes it easy to follow a series of log entries in
      concurrent code
    - ability to produce a chrome://tracing-compatible trace file,
      either via an env variable or a `zrepl pprof` subcommand
        - this is not a CPU profile, we already have go pprof for that
        - but it is very useful to visually inspect where the
          replication / snapshotter / pruner spends its time
          ( fixes #307 )

rpc/dataconn/stream/stream.go

@@ -7,6 +7,7 @@ import (
 	"io"
 	"net"
 	"strings"
+	"sync"
 	"sync/atomic"
 	"unicode/utf8"
 
@@ -80,9 +81,13 @@ func doWriteStream(ctx context.Context, c *heartbeatconn.Conn, stream io.Reader,
 		err error
 	}
 
+	var wg sync.WaitGroup
+	defer wg.Wait()
 	reads := make(chan read, 5)
 	var stopReading uint32
+	wg.Add(1)
 	go func() {
+		defer wg.Done()
 		defer close(reads)
 		for atomic.LoadUint32(&stopReading) == 0 {
 			buffer := bufpool.Get(1 << FramePayloadShift)