cmd: Add 'caddy status' command and core StatusReporter interface

[pauloappbr]

Assistance Disclosure

I used an LLM (Gemini) as a pair-programming assistant to brainstorm the architecture, design the StatusReporter interface, and generate the initial Go code structure. I then compiled, debugged, and manually verified the correctness of the code locally to ensure it meets Caddy's standards.

Resolves Phase 1 of #7453.

As discussed in the issue, this PR introduces the core architecture and the CLI command to view the running status of a Caddy instance without duplicating /debug/pprof logic, setting the ground for app-specific metrics.

What this PR does:

1. Extensibility: Introduces the StatusReporter interface in the core caddy package. Any App or Module can implement this interface to safely expose its internal status snapshot.
2. Admin API: Adds the GET /status endpoint. It loops through the ActiveContext() loaded apps, aggregating basic process data (Uptime, Memory, Goroutines, Version) and dynamically appending any app-specific status if the app implements StatusReporter.
3. CLI: Adds the caddy status command. By default, it outputs a human-readable summary of the JSON payload. The --json flag is available for raw output.

Future Work (Phase 2):

Implementing the StatusReporter interface specifically inside the http app to track active requests, latency, and the dot-matrix scoreboard, leveraging existing Prometheus hooks.

Demo:

Standard Output:

Caddy v2.x.x
Status: Running
Uptime: 14m 32s

Running apps: http, tls

Memory
Allocated: 4 MB
System: 19 MB
Goroutines: 18

JSON Output (caddy status --json):

{
  "apps": {
    "http": {"status":"running"},
    "tls": {"status":"running"}
  },
  "goroutines": 18,
  "memory": {
    "allocated_bytes": 4071992,
    "system_bytes": 19749128
  },
  "uptime_secs": 872.4,
  "version": "v2.x.x"
}

[mholt]

Thanks for the PR! I'm excited to clone this down and kick it around a bit.