Local span perf improvements

[jtescher]

Currently span data recording performance is capped because all changes are synchronized behind a mutex. This patch changes the span API to take mutable references instead, and only requires the use of mutexes when setting values through a shared reference to the active span in the context API.

results: 19-47% performance improvement

start-end-span/always-sample
                        time:   [321.64 ns 323.85 ns 326.40 ns]
                        change: [-48.064% -47.628% -47.169%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 8 outliers among 100 measurements (8.00%)
  6 (6.00%) high mild
  2 (2.00%) high severe
start-end-span/never-sample
                        time:   [219.28 ns 220.25 ns 221.29 ns]
                        change: [-34.537% -33.419% -32.409%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 10 outliers among 100 measurements (10.00%)
  8 (8.00%) high mild
  2 (2.00%) high severe

start-end-span-4-attrs/always-sample
                        time:   [895.10 ns 904.83 ns 916.50 ns]
                        change: [-31.319% -30.690% -30.038%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 6 outliers among 100 measurements (6.00%)
  3 (3.00%) high mild
  3 (3.00%) high severe
start-end-span-4-attrs/never-sample
                        time:   [249.18 ns 250.59 ns 252.33 ns]
                        change: [-30.774% -29.948% -29.168%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 13 outliers among 100 measurements (13.00%)
  8 (8.00%) high mild
  5 (5.00%) high severe

start-end-span-8-attrs/always-sample
                        time:   [1.5984 us 1.6098 us 1.6226 us]
                        change: [-27.169% -25.705% -24.416%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 6 outliers among 100 measurements (6.00%)
  3 (3.00%) high mild
  3 (3.00%) high severe
start-end-span-8-attrs/never-sample
                        time:   [287.27 ns 288.77 ns 290.62 ns]
                        change: [-24.716% -23.535% -22.466%] (p = 0.00 < 0.05)
                        Performance has improved.

start-end-span-all-attr-types/always-sample
                        time:   [1.2813 us 1.2911 us 1.3018 us]
                        change: [-25.382% -24.399% -23.483%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 3 outliers among 100 measurements (3.00%)
  1 (1.00%) high mild
  2 (2.00%) high severe
start-end-span-all-attr-types/never-sample
                        time:   [263.15 ns 264.19 ns 265.32 ns]
                        change: [-31.910% -30.959% -30.085%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 7 outliers among 100 measurements (7.00%)
  2 (2.00%) high mild
  5 (5.00%) high severe

start-end-span-all-attr-types-2x/always-sample
                        time:   [2.2835 us 2.2974 us 2.3119 us]
                        change: [-20.158% -19.254% -18.254%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 5 outliers among 100 measurements (5.00%)
  3 (3.00%) high mild
  2 (2.00%) high severe
start-end-span-all-attr-types-2x/never-sample
                        time:   [312.36 ns 313.86 ns 315.61 ns]
                        change: [-27.570% -26.865% -26.059%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 6 outliers among 100 measurements (6.00%)
  2 (2.00%) high mild
  4 (4.00%) high severe

[frigus02]

This looks like a good improvement. Nice work!

[djc]

Nice improvement!