Phase 3: dashboards and alerting

Saved, shareable multi-panel dashboards (table/line/bar/single-stat
panels via gridstack + uPlot, global + per-panel time range, JSON
export/import) and threshold/absence alert rules with an
ok/pending/firing evaluator and webhook/Slack/PagerDuty delivery.

- New /metadata component: Postgres control-plane store for dashboards,
  panels, notification targets, alert rules/state, and delivery log --
  see docs/phase-3-dashboard-design.md for why ClickHouse's MergeTree
  family isn't a fit for this access pattern (needs real row-level
  locking and read-your-writes consistency).
- api/internal/dashboards: dashboard/panel CRUD, pure -- panel query
  execution stays client-side, reusing the existing /query endpoint.
- New /alerting service: rule/target CRUD, a ticker-driven evaluator
  (claim-then-evaluate concurrency control, transactional-outbox
  delivery, query errors and threshold zero-rows never coerced into a
  false transition) and webhook/Slack/PagerDuty delivery with
  retry/backoff. See docs/phase-3-alerting-design.md for the full
  state-machine design and the four correctness properties it
  implements.
- web: /dashboards and /alerts UIs; cli: sentryctl dashboards/alerts
  list/get/apply, seeding a future Terraform provider's JSON contract.
- hack/alert-load-test: 500 rules against real ClickHouse data, real
  measured results in docs/phase-3-runbook.md.

Five real bugs found by actually running this against a live stack
(documented in the runbook, not just fixed silently): a latent Phase 2
bug where ClickHouse rejected the timestamp format used for
earliest=/latest= queries; a "now" literal token injected into query
text; a GridStack/uPlot layout-timing race; JS's Date.parse being too
lenient to use as a timestamp-detection heuristic; a rule's "enabled"
field silently defaulting to false when omitted; and the evaluator's
claim-batch-size and worker-pool-concurrency defaulting to the same
value, causing 500 concurrently-due rules to take 125s to cycle through
instead of the configured 60s.

cli/cmd/sentryctl/cmd_dashboards.go

@@ -0,0 +1,55 @@
+package main
+
+import (
+	"fmt"
+	"io"
+	"os"
+)
+
+func cmdDashboards(args []string, stdout, stderr io.Writer) int {
+	if len(args) == 0 {
+		fmt.Fprintln(stderr, "sentryctl dashboards: expected a subcommand (list, get, apply)")
+		return 1
+	}
+	apiURL, rest := extractAPIFlag(args[1:], os.Getenv)
+
+	switch args[0] {
+	case "list":
+		return httpGetJSON(apiURL, "/dashboards", stdout, stderr)
+	case "get":
+		if len(rest) == 0 {
+			fmt.Fprintln(stderr, "sentryctl dashboards get: missing dashboard id")
+			return 1
+		}
+		return httpGetJSON(apiURL, "/dashboards/"+rest[0], stdout, stderr)
+	case "apply":
+		if len(rest) == 0 {
+			fmt.Fprintln(stderr, "sentryctl dashboards apply: missing file path")
+			return 1
+		}
+		// The import endpoint consumes exactly the shape GET
+		// /dashboards/{id}/export produces and the web UI's Export JSON
+		// button downloads -- one JSON contract, three call sites.
+		return httpPostFileJSON(apiURL, "/dashboards/import", rest[0], stdout, stderr)
+	default:
+		fmt.Fprintf(stderr, "sentryctl dashboards: unknown subcommand %q (want list, get, apply)\n", args[0])
+		return 1
+	}
+}
+
+// extractAPIFlag pulls an optional --api <url> out of args, resolving
+// the default the same way parsePingArgs/parseQueryArgs do, and returns
+// the remaining positional args. Shared by dashboards and alerts since
+// both take an optional --api/--alerting-api override the same way.
+func extractAPIFlag(args []string, env func(string) string) (apiURL string, rest []string) {
+	apiURL = resolveAPIURL(env)
+	for i := 0; i < len(args); i++ {
+		if args[i] == "--api" && i+1 < len(args) {
+			apiURL = args[i+1]
+			i++
+			continue
+		}
+		rest = append(rest, args[i])
+	}
+	return apiURL, rest
+}