perf(p2p-13): optimize the data-plane fast path (#671)

[jlon]

Summary

• Land the 0ms-critical data-plane fast path after the redirect path is stable.
• Keep rollout posture unchanged; this PR is performance-only and excludes adaptive logic and telemetry.

Design

• Optimize the first redirected chunk without changing protocol semantics or fallback behavior.
• Isolate fast-path buffer and first-chunk work from adaptive heuristics and validation assets.

Implementation

• Avoid zeroing response buffers and stream the first data-plane chunk earlier.
• Shape the data-plane read path for lower latency, with remote-reader adjustments only where required.

Verification

• Run the production gate for redirected reads.
• Check 0ms benchmark sanity and targeted redirected-read tests.

[chatgpt-codex-connector]

Codex usage limits have been reached for code reviews. Please check with the admins of this repo to increase the limits by adding credits.
Credits must be used to enable repository wide code reviews.