fix: close SSRF in stream proxies and fix dotenv load order

Security:
- Fixed unauthenticated SSRF in /api/stream/proxy and /api/stream/ts-proxy
  - proxy allowlist had escape hatch via pathname.endsWith('.m3u8') bypass
  - ts-proxy had no destination validation at all
  - Both could fetch internal services (127.0.0.1, 169.254.169.254)
- Added streamProxy.helper.js with domain allowlist, private IP blocking,
  and http(s)-only enforcement applied to both endpoints
- Removed dead megaplay-1.buzz; added vid-tube.site, norami.top,
  kryntal.top, akirax.buzz to allowlist
- Added SECURITY.md with vulnerability reporting policy

Bug fix:
- Fixed dotenv.config() loading after static imports -- helpers reading
  process.env at top level never received .env values. Changed to
  import 'dotenv/config' as first import

Config:
- Added STREAM_PROXY_DOMAINS env var for custom allowlist
- Added STREAM_PROXY_REFERER env var for custom CDN Referer

SECURITY.md

@@ -0,0 +1,54 @@
+# Security Policy
+
+## Supported versions
+
+| Version | Supported |
+| ------- | --------- |
+| 2.4.x   | ✅        |
+| < 2.4   | ❌        |
+
+Fixes land on the latest release. If you are running an older version, upgrade before reporting — the issue may already be resolved.
+
+## Reporting a vulnerability
+
+**Please do not open a public issue or pull request for a security problem.** Public reports expose every running instance until a fix ships.
+
+Report privately instead:
+
+1. **Preferred — GitHub private vulnerability reporting.** Use the **Report a vulnerability** button on the [Security tab](https://github.com/Shineii86/AniKotoAPI/security). This keeps the discussion private until an advisory is published.
+2. **Fallback — Telegram.** If private reporting is unavailable, message [@Shineii86](https://telegram.me/Shineii86) asking for a private channel. Send only a one-line description there, not the full details.
+
+## What to include
+
+The more of this you can provide, the faster a fix can be confirmed:
+
+- The affected endpoint or module, and the version or commit you tested.
+- What an attacker can achieve — reading data, changing state, exhausting resources.
+- Minimal steps to reproduce, ideally a single request.
+- Whether authentication or any particular deployment setup is required.
+- A suggested fix, if you have one.
+
+## What happens next
+
+- Reports are acknowledged as soon as the maintainer sees them.
+- Once confirmed, a fix is prepared privately and released, and a GitHub advisory is published crediting the reporter unless anonymity is requested.
+- Reporters are asked to hold public details until the fix is released.
+
+This project is maintained by one person as unpaid work. Please read the timing above as intent rather than a guaranteed schedule.
+
+## Scope
+
+In scope — anything that lets someone abuse a deployed instance:
+
+- Server-side request forgery, or any way to make the server fetch or reach destinations it should not.
+- Injection, path traversal, or unsafe handling of caller-supplied input.
+- Leakage of environment variables, credentials, or internal configuration.
+- Denial of service reachable through a single unauthenticated request.
+- Vulnerable dependencies with a practical path to exploitation in this codebase.
+
+Out of scope:
+
+- Availability, downtime, or changes to the upstream anime sites this API scrapes.
+- Rate limiting or abuse controls on those upstream sites.
+- Findings that require an already-compromised host or shell access to the server.
+- Reports produced by automated scanners with no demonstrated impact.