resources: Update state checks

The mgmt graph depends on state tracking to eliminate redundant pokes.
With the Watch loop now able to produce events quickly, it should no
longer play a part in determining the vertex state. This simplifies the
resource API as well!

test/shell/t5b.sh

@@ -0,0 +1,7 @@
+#!/bin/bash -e
+
+# should take slightly more than 35s, but fail if we take 45s)
+timeout --kill-after=45s 40s ./mgmt run --yaml t5.yaml --converged-timeout=5 --no-watch --tmp-prefix &
+pid=$!
+wait $pid	# get exit status
+exit $?

test/shell/t5b.yaml

@@ -0,0 +1,84 @@
+---
+graph: mygraph
+comment: simpler exec fan in to fan out example to demonstrate optimization
+resources:
+  exec:
+  - name: exec1
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec2
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec3
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec4
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec5
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+edges:
+- name: e1
+  from:
+    kind: exec
+    name: exec1
+  to:
+    kind: exec
+    name: exec3
+- name: e2
+  from:
+    kind: exec
+    name: exec2
+  to:
+    kind: exec
+    name: exec3
+- name: e3
+  from:
+    kind: exec
+    name: exec3
+  to:
+    kind: exec
+    name: exec4
+- name: e4
+  from:
+    kind: exec
+    name: exec3
+  to:
+    kind: exec
+    name: exec5