… (DRIVER-201)

Addresses the initial-contact-endpoints aspect of DRIVER-201.

Problem: with RESOLVE_CONTACT_POINTS=false (the default), a contact
point hostname was stored as a single unresolved InetSocketAddress.
At connection time the load-balancing query plan contained exactly one
Node per hostname, so only the first IP returned by DNS was ever tried.
If that IP was non-responsive the driver raised AllNodesFailedException
with no fallback to other IPs the hostname might resolve to.

Solution (per @dkropachev's architectural direction):
- Deprecate DefaultDriverOption.RESOLVE_CONTACT_POINTS. Contact points
  are now always kept as unresolved hostnames (resolve=false is hardcoded
  in SessionBuilder), deferring DNS expansion to connection time.
- Add MetadataManager.getResolvedContactPoints(): for each contact point
  backed by an unresolved hostname it calls InetAddress.getAllByName() to
  expand the hostname to all known IPs, creating a synthetic DefaultNode
  for each IP. Already-resolved or non-InetSocketAddress endpoints pass
  through unchanged.
- LoadBalancingPolicyWrapper now calls getResolvedContactPoints() instead
  of getContactPoints() in newQueryPlan() (BEFORE/DURING_INIT states) and
  newControlReconnectionQueryPlan(), so the query plan contains one node
  per resolved IP and the driver naturally falls back to the next IP when
  one is unreachable.

Tests:
- 4 new MetadataManagerTest cases covering null state, already-resolved
  passthrough, single-hostname expansion, and multi-endpoint expansion.
- LoadBalancingPolicyWrapperTest updated to stub getResolvedContactPoints().
- New MockResolverIT.should_connect_when_first_dns_entry_is_non_responsive
  integration test: first DNS entry is a non-existent IP, session must
  open successfully against the remaining real IPs.

…VER-201)

newControlReconnectionQueryPlan() now creates copies of the original
contact-point nodes (with their unresolved hostname endpoints) instead
of synthetic nodes with resolved IPs. This ensures the control channel
carries the hostname endpoint, which is preserved in metadata after
topology refresh.

DNS expansion for connection fallback is handled by ChannelFactory
(PR scylladb#890), so the control-reconnection path does not need to inject
resolved-IP nodes into the query plan.

Also adds getContactPoints() stub back to LoadBalancingPolicyWrapperTest
so tests that cover the control-reconnect path continue to pass.