refactor(udp): remove useless Socket::send return value

[mxinden]

Socket::send returns the number of Datagrams sent. Given that it only accepts a single Datagram to be send, this return value is at best useless and in the worst case confusing. It is ignored at all call sites.

---

Introduced in #1604. My bad. Sorry for the noise.

[github-actions]

Benchmark results

Performance differences relative to 2750423.

• coalesce_acked_from_zero 1+1 entries
  time: [254.14 ns 255.19 ns 256.82 ns]
  change: [-0.7895% -0.4172% +0.0780%] (p = 0.04 < 0.05)
  Change within noise threshold.

• coalesce_acked_from_zero 3+1 entries
  time: [290.21 ns 290.94 ns 291.77 ns]
  change: [-0.5469% -0.2358% +0.0742%] (p = 0.13 > 0.05)
  No change in performance detected.

• coalesce_acked_from_zero 10+1 entries
  time: [290.18 ns 291.47 ns 292.92 ns]
  change: [-1.1413% -0.5410% -0.0434%] (p = 0.05 > 0.05)
  No change in performance detected.

• coalesce_acked_from_zero 1000+1 entries
  time: [267.32 ns 267.44 ns 267.60 ns]
  change: [-0.9423% -0.3758% +0.2188%] (p = 0.22 > 0.05)
  No change in performance detected.

• RxStreamOrderer::inbound_frame()
  time: [125.73 ms 125.90 ms 126.15 ms]
  change: [-1.2696% -1.0398% -0.7892%] (p = 0.00 < 0.05)
  Change within noise threshold.

• Run multiple transfers with varying seeds
  time: [189.43 ms 190.01 ms 190.58 ms]
  change: [-1.5510% -1.1153% -0.6858%] (p = 0.00 < 0.05)
  Change within noise threshold.

• Run multiple transfers with the same seed
  time: [189.74 ms 190.11 ms 190.51 ms]
  change: [-1.5651% -1.2710% -0.9860%] (p = 0.00 < 0.05)
  Change within noise threshold.

⬇️ Download logs