Use multiply shift instead of u128 division to 2x encoding speed

[jackkleeman]

With some help from the quickdiv library, I have found the multiply and shift that is equivalent to a u128 division by 62^10. We currently do that division twice and its super super slow (__udivti3). For u64 the compiler already turns the divisions into a multiply and shift which is why the use of mostly-u64-divisions in this library is so effective, but this is not the case for u128 apparently.

encode/standard_new_fixed
                        time:   [32.628 ns 32.801 ns 33.085 ns]
                        change: [-41.146% -39.914% -39.003%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 4 outliers among 100 measurements (4.00%)
  2 (2.00%) high mild
  2 (2.00%) high severe
encode/standard_new_random
                        time:   [30.778 ns 31.021 ns 31.250 ns]
                        change: [-46.061% -45.407% -44.802%] (p = 0.00 < 0.05)
                        Performance has improved.
encode/standard_bytes_fixed
                        time:   [21.184 ns 21.255 ns 21.352 ns]
                        change: [-50.594% -49.762% -48.248%] (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
encode/standard_bytes_random
                        time:   [23.588 ns 23.664 ns 23.744 ns]
                        change: [-49.891% -49.451% -48.838%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 4 outliers among 100 measurements (4.00%)
  3 (3.00%) high mild
  1 (1.00%) high severe
encode/standard_buf_fixed
                        time:   [36.390 ns 36.488 ns 36.590 ns]
                        change: [-35.510% -35.272% -34.976%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 3 outliers among 100 measurements (3.00%)
  2 (2.00%) high mild
  1 (1.00%) high severe
encode/standard_buf_random
                        time:   [23.543 ns 23.669 ns 23.813 ns]
                        change: [-48.243% -47.933% -47.579%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 4 outliers among 100 measurements (4.00%)
  1 (1.00%) high mild
  3 (3.00%) high severe
encode/alternative_new_fixed
                        time:   [32.984 ns 33.088 ns 33.180 ns]
                        change: [-39.498% -39.270% -39.035%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high severe
encode/alternative_new_random
                        time:   [30.405 ns 30.618 ns 30.859 ns]
                        change: [-45.348% -44.161% -42.669%] (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
encode/alternative_bytes_fixed
                        time:   [21.306 ns 21.358 ns 21.419 ns]
                        change: [-50.234% -50.088% -49.940%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
  1 (1.00%) high mild
  1 (1.00%) high severe
encode/alternative_bytes_random
                        time:   [24.073 ns 24.297 ns 24.546 ns]
                        change: [-49.165% -48.685% -48.177%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 4 outliers among 100 measurements (4.00%)
  2 (2.00%) high mild
  2 (2.00%) high severe
encode/alternative_buf_fixed
                        time:   [36.533 ns 36.621 ns 36.718 ns]
                        change: [-35.979% -35.560% -35.046%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
  1 (1.00%) high mild
  1 (1.00%) high severe
encode/alternative_buf_random
                        time:   [23.507 ns 23.579 ns 23.658 ns]
                        change: [-49.875% -49.461% -49.169%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild