Run the daemon worker on the same CPU as the client

On a system with multiple CPUs, running Nix operations through the
daemon is significantly slower than "direct" mode:

$ NIX_REMOTE= nix-instantiate '<nixos>' -A system
real    0m0.974s
user    0m0.875s
sys     0m0.088s

$ NIX_REMOTE=daemon nix-instantiate '<nixos>' -A system
real    0m2.118s
user    0m1.463s
sys     0m0.218s

The main reason seems to be that the client and the worker get moved
to a different CPU after every call to the worker.  This patch adds a
hack to lock them to the same CPU.  With this, the overhead of going
through the daemon is very small:

$ NIX_REMOTE=daemon nix-instantiate '<nixos>' -A system
real    0m1.074s
user    0m0.809s
sys     0m0.098s

src/libstore/build.cc

@@ -7,6 +7,7 @@
 #include "local-store.hh"
 #include "util.hh"
 #include "archive.hh"
+#include "affinity.hh"
 
 #include <map>
 #include <sstream>
@@ -366,6 +367,8 @@ void Goal::trace(const format & f)
 /* Common initialisation performed in child processes. */
 static void commonChildInit(Pipe & logPipe)
 {
+    restoreAffinity();
+
     /* Put the child in a separate session (and thus a separate
        process group) so that it has no controlling terminal (meaning
        that e.g. ssh cannot open /dev/tty) and it doesn't receive
@@ -568,6 +571,7 @@ static void runSetuidHelper(const string & command,
             args.push_back(0);
 
             restoreSIGPIPE();
+            restoreAffinity();
 
             execve(program.c_str(), (char * *) &args[0], 0);
             throw SysError(format("executing `%1%'") % program);