HydraCache 0.59.0

HydraCache 0.59.0

0.59.0 ships the first deployable networked daemon grid. Member-mode
hydracache-server now wires the shipped discovery, raft metadata runtime, and
cluster transport adapters into one live membership authority, closing the
0.57 Management Center live-grid gap and the TD-0008 daemon-hosting gap.

The release integrates shipped raft-rs mechanics; it does not introduce a new
consensus algorithm. Local and client roles remain modeled, and the embedded
cache fast path is unchanged.

Highlights

• Added the networked member stack for hydracache-server: chitchat discovery,
  durable/networked RaftMetadataRuntime, the admission bridge, and the axum
  cluster transport.
• Replaced the in-process-only member status path with NetworkedGridHandle
  backed by the same raft runtime that owns membership.
• Added loopback multi-daemon evidence for cluster formation, leader election,
  leader loss, re-election, and no stale leader in the live status surface.
• Made the Management Center and admin status path report a real live source
  for member-mode daemon grids.
• Re-pointed the 0.58 multi-node soak disclosure toward the networked daemon
  grid while keeping partition and slow-disk chaos as external-injector
  evidence.

Verification

Focused gates for the networked daemon stack:

cargo test -p hydracache-cluster-raft --locked
cargo test -p hydracache-server --test grid_host --locked
$env:HYDRACACHE_RUN_NETWORKED_DAEMON_E2E='1'
cargo test -p hydracache-server --test grid_host multi_node --locked -- --nocapture
cargo xtask verify

Scope Boundary

0.59.0 deliberately stops at loopback-grade networked hosting. Cluster
transport TLS termination, peer auth, persistent node identity, dynamic raft
membership, and stronger CI coverage are named follow-up hardening work in
0.60.0.