refactor!: rebrand CCAM to AgentHub

Clean break, no compatibility alias. Renames the product name, the CLI
binary (`ccam` -> `agenthub`), the env var prefix (`CCAM_` ->
`AGENTHUB_`), the Prometheus metric namespace (`ccam_*` ->
`agenthub_*`), the tmux session prefix (`ccam-lane-<id>` ->
`agenthub-lane-<id>`), the per-repo runtime profile directory
(`.ccam/profile` -> `.agenthub/profile`), the lanes state root
(`~/.claude/ccam-lanes` -> `~/.claude/agenthub-lanes`), the bundled
plugins (`plugins/ccam*` -> `plugins/agenthub*`) and their slash
commands (`/ccam-doctor` -> `/agenthub-doctor`, `/ccam:*` ->
`/agenthub:*`).

Because there is no alias, an existing install has to be updated by
hand: rename its `.ccam/profile` directory and any `CCAM_*` variables,
and reinstall the plugin under its new name.

Two other name families are deliberately left alone here, since they are
not just strings: the `Claude-Code-Monitor` repository/marketplace name,
and the `agent-dashboard` npm package plus its `~/.claude/agent-dashboard`
runtime directory, which holds the live database and needs a migration
rather than a rename.

Verified: server 1406/1406, client 459/459, `grep -i ccam` clean across
tracked files, `openapi.yaml` regenerated (95 paths, unchanged count).

server/lib/lane-runtime.js

@@ -4,23 +4,23 @@
  * data-isolation lifecycle (A2): `provisionLane`/`resetLaneData`/`removeLaneData`
  * seed `.env`, create/drop the lane's own database, and run `bootstrap`/
  * `migrate`/`seed` at the points Shipyard's `lane-bootstrap.sh`/`lane-up.sh`/
- * `lane-reset.sh`/`lane-remove.sh` do — CCAM's version of "ports came from A1,
+ * `lane-reset.sh`/`lane-remove.sh` do — AgentHub's version of "ports came from A1,
  * everything else a lane needs to run its own stack comes from here."
  *
  * Two properties shape everything here.
  *
  * Services are FULLY DETACHED (see `harness_spawn` in lane-profile.js), so a
  * lane's stack outlives both the hook that started it and the dashboard itself —
- * restarting or updating CCAM must never kill work in progress.
+ * restarting or updating AgentHub must never kill work in progress.
  *
- * Liveness is COMPUTED, never stored. Whether a stack is up is not a fact CCAM
+ * Liveness is COMPUTED, never stored. Whether a stack is up is not a fact AgentHub
  * controls: a process dies to OOM, to a stray `kill`, to a reboot. A cached
  * "running" flag would be wrong from that moment until something noticed, so
  * `runtimeFacts` re-derives it from pid files and port probes on every read. The
  * payoff is that adopting a stack after a dashboard restart needs no code at all.
  *
  * This module writes `slot` and `ports` and nothing else on the lane row. It never
- * writes `stage`, `status` or `notes`: in CCAM those describe the AGENT's work, not
+ * writes `stage`, `status` or `notes`: in AgentHub those describe the AGENT's work, not
  * the stack's state, and conflating "a server is listening" with "an agent is
  * running" would corrupt lane liveness. Boot failures live in `last-error.json`
  * and surface through `runtimeFacts`.
@@ -58,7 +58,9 @@ function requireProfile(lane) {
   const profile = resolveProfile(lane);
   if (!profile) {
     throw Object.assign(
-      new Error(`lane has no .ccam/profile — looked in: ${profileSearchPaths(lane).join(", ")}`),
+      new Error(
+        `lane has no .agenthub/profile — looked in: ${profileSearchPaths(lane).join(", ")}`
+      ),
       { code: "ENOPROFILE", searched: profileSearchPaths(lane) }
     );
   }
@@ -299,8 +301,8 @@ async function resetLaneData(lane, profile, options = {}) {
  * Drop a managed lane's database and its `_test` sibling before the rest of
  * removal tears down the worktree and state directory.
  *
- * A no-op for an adopted lane — its data was never CCAM's to create, so it is
- * never CCAM's to destroy either, the same invariant `assertDestroyable`
+ * A no-op for an adopted lane — its data was never AgentHub's to create, so it is
+ * never AgentHub's to destroy either, the same invariant `assertDestroyable`
  * enforces for the worktree itself. Best-effort per database: mirrors
  * Shipyard's `dropdb --if-exists ... || true` — a failed drop is logged, not
  * thrown, because it must never block removing the lane's dashboard record.