This commit extracts the fingerprint logic from the csi instance manager
into its own struct.

This change is to facilitate reusing the csimanager to also manage the
node-local CSI functionality, as it is the easiest place for us to
guaruntee health checking and to provide additional visibility into the
running operations through the fingerprinter mechanism and goroutine.

Implements a fake version of NodeStageVolume as a dependency of
implementing the client.NodeStageVolume request

This commit introduces a nomad model for interacting with CSI
VolumeCapabilities as a pre-requisite for implementing NodeStageVolume
and NodeMountVolume correctly.

These fields have a few special characteristics that I've tried to model
here - specificially, we make a basic attempt to avoid printing data
that should be redacted during debug logs (additional mount flags), and
also attempt to make debuggability of other integer fields easier by
implementing the fmt.Stringer and fmt.GoStringer interfaces as
necessary.

We do not currnetly implement a CSI Protobuf -> Nomad implementation
transformation as this is currently not needed by any used RPCs.

state_store: change claim counts

state_store: get volumes by all, by driver

state_store: process volume claims

state_store: csi volume register error on update

This commit introduces two new fields to the basic output of `nomad
node status <node-id>`.

1) "CSI Controllers", which displays the names of registered controller
plugins.

2) "CSI Drivers", which displays the names of registered CSI Node
plugins.

However, it does not implement support for verbose output, such as
including health status or other fingerprinted data.