feat(auth): add PAT auth, scoped REST/MCP access, CSRF, and tenant isolation

Introduce Personal Access Tokens (nzp_*) as a stateless auth path alongside
JWT, gated per-endpoint by a scope middleware (nezha:{resource}:{verb}) with
fail-closed empty-scope defaults and a server-id whitelist. Self-management
endpoints (profile, api-tokens, oauth2 bind, refresh-token) explicitly reject
PATs to block privilege-escalation chains. A revoke registry tears down active
long-lived connections (terminal, fm, ws, transfer, mcp) the moment a PAT is
deleted, with a tombstone closing the revoke->register race.

Add an MCP endpoint that proxies tool calls (exec, fs read/write/delete,
transfer) to agents over gRPC, guarded by origin/DNS-rebinding checks, a
per-token rate limiter, audit logging, and a kill switch. Serialize all
sends through the IOStream wrapper to honour grpc-go's concurrency contract.

Add CSRF double-submit protection on unsafe cookie-authenticated methods,
exempting authenticated PAT requests by context identity (not a forgeable
Authorization header). Apply visibility/whitelist filtering consistently
across list, get-by-id, and mutate paths to enforce tenant isolation.

Migrate legacy mcp:* scopes: rewrite read/exec to nezha:* equivalents and
drop dangerous write/delete/wildcard grants.

Co-authored-by: cloudcode <cloudcode@users.noreply.github.com>

cmd/dashboard/controller/mcp_ratelimit_prune_test.go

@@ -0,0 +1,62 @@
+package controller
+
+import (
+	"testing"
+	"time"
+)
+
+// The per-token limiter map had no eviction: every distinct token ID ever
+// seen left a permanent entry. A user churning PATs (create/use/delete in a
+// loop) grows the map without bound. Allow must opportunistically prune
+// windows idle past the minute bucket so memory stays proportional to the
+// active token set, not the historical one.
+func TestMCPRateLimiter_PrunesStaleTokenWindows(t *testing.T) {
+	rl := newMCPRateLimiter(10, 120)
+
+	stale := time.Now().Add(-10 * time.Minute)
+	for i := uint64(1); i <= 500; i++ {
+		rl.mu.Lock()
+		rl.perToken[i] = &tokenWindow{
+			secBucketStart: stale,
+			minBucketStart: stale,
+		}
+		rl.mu.Unlock()
+	}
+
+	// A fresh request triggers a prune sweep of idle windows.
+	if !rl.Allow(99999) {
+		t.Fatal("fresh token must be allowed")
+	}
+
+	rl.mu.Lock()
+	size := len(rl.perToken)
+	rl.mu.Unlock()
+
+	// Only the just-active token (99999) should remain; the 500 stale ones
+	// must have been evicted.
+	if size > 1 {
+		t.Fatalf("stale token windows were not pruned: map still holds %d entries", size)
+	}
+}
+
+// Pruning must NOT evict tokens that are still within their active window,
+// otherwise an in-flight client loses its accumulated count and effectively
+// resets its budget.
+func TestMCPRateLimiter_KeepsActiveTokenWindows(t *testing.T) {
+	rl := newMCPRateLimiter(10, 120)
+
+	if !rl.Allow(1) {
+		t.Fatal("token 1 must be allowed")
+	}
+	if !rl.Allow(2) {
+		t.Fatal("token 2 must be allowed")
+	}
+
+	rl.mu.Lock()
+	size := len(rl.perToken)
+	rl.mu.Unlock()
+
+	if size != 2 {
+		t.Fatalf("active token windows must be retained, got %d entries", size)
+	}
+}