Update 3 links to the hierachy of the new-docs site

Signed-off-by: Bernhard Kaindl <[email protected]>

doc/content/toolstack/features/DR/index.md

@@ -2,7 +2,7 @@
 title = "Disaster Recovery"
 +++
 
-The [HA](../HA/HA.html) feature will restart VMs after hosts have failed, but what
+The [HA](HA) feature will restart VMs after hosts have failed, but what
 happens if a whole site (e.g. datacenter) is lost? A disaster recovery
 configuration is shown in the following diagram:
 

doc/content/toolstack/features/snapshots/index.md

@@ -8,7 +8,7 @@ Snapshots represent the state of a VM, or a disk (VDI) at a point in time. They
 - experiments (take snapshot, try something, revert back again)
 - golden images (install OS, get it just right, clone it 1000s of times)
 
-Read more about [the Snapshot APIs](../../xen-api/snapshots.html).
+Read more about [the Snapshot APIs](../../../xen-api/topics/snapshots).
 
 Disk snapshots
 ==============

doc/content/xenopsd/architecture/_index.md

@@ -17,7 +17,7 @@ Managing a VM means:
 - providing updates to clients when things change (reboots, console becomes
   available, guest agent says something etc).
 
-For a full list of features, consult the [features list](features.html).
+For a full list of features, consult the [feature list](../features).
 
 Each Xenopsd instance has a unique name on the host. A typical name is