perf(proxy): add sync.Pool for UDP buffers

[LaurenceJJones]

Community Contribution License Agreement

By creating this pull request, I grant the project maintainers an unlimited,
perpetual license to use, modify, and redistribute these contributions under any terms they
choose, including both the AGPLv3 and the Fossorial Commercial license terms. I
represent that I have the right to grant this license for all contributed content.

Description

• Add udpBufferPool for reusable 65507-byte UDP packet buffers
• Add getUDPBuffer() and putUDPBuffer() helper functions
• Clear buffer contents before returning to pool to prevent data leakage
• Apply pooling to both main handler buffer and per-client goroutine buffers
• Reduces GC pressure from frequent large allocations during UDP proxying

How to test?

Internal code changes.

[strausmann]

We can confirm the UDP buffer allocation issue described in this PR from our production environment.

On an affected Newt instance (details in #268), we observed 11+ simultaneous UDP connections to our DNS resolver, each with its own file descriptor — no reuse:

UDP: fd=86   -> dns-resolver:853
UDP: fd=479  -> dns-resolver:853
UDP: fd=1057 -> dns-resolver:853
UDP: fd=1232 -> dns-resolver:853
UDP: fd=2749 -> dns-resolver:853

Under high TCP forwarding load (SMTP targets with health checks), DNS lookups are very frequent, which amplifies the buffer allocation issue significantly. In proxy/manager.go, every new TCP connection in handleTCPProxy triggers a net.Dial which performs DNS resolution — and under our workload we saw 246 TCP connections within a 500-line log window.

The sync.Pool approach in this PR would directly address what we're seeing on the UDP side. Combined with a fix for the TCP connection leak (#268), it would significantly reduce the memory and FD growth we're experiencing.

Would love to see this merged.