impl shr in Arm assembly

[dvdplm]

As a follow-up to the conversation in #753, here's an implementation of right shift using Arm assembly. I don't suggest we ever merge this, but it does give us a way to measure the constant-time-ness of alternative impls, such as the one suggested in #753.

cc @erik-3milabs

On my mac the benchmarks numbers are:

right shift/shr, U2048  time:   [179.15 ns 180.23 ns 181.56 ns]
                        change: [+0.7871% +1.5747% +2.4028%] (p = 0.00 < 0.05)
                        Change within noise threshold.
right shift/shr, U2048, asm
                        time:   [30.854 ns 31.279 ns 31.621 ns]
                        change: [-3.8200% +0.3339% +5.0328%] (p = 0.88 > 0.05)
                        No change in performance detected.
Found 12 outliers among 100 measurements (12.00%)
  12 (12.00%) low mild
right shift/shr, U2048, zero
                        time:   [179.11 ns 180.27 ns 181.68 ns]
                        change: [-1.1264% +0.1229% +1.3662%] (p = 0.86 > 0.05)
                        No change in performance detected.
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high severe
right shift/shr, U2048, zero, asm
                        time:   [30.943 ns 31.393 ns 31.753 ns]
                        change: [-2.1681% +2.2669% +6.7396%] (p = 0.32 > 0.05)
                        No change in performance detected.
Found 13 outliers among 100 measurements (13.00%)
  13 (13.00%) low mild
right shift/shr, U2048, mid
                        time:   [189.42 ns 190.02 ns 190.62 ns]
                        change: [-1.1437% -0.2175% +0.6928%] (p = 0.64 > 0.05)
                        No change in performance detected.
right shift/shr, U2048, mid, asm
                        time:   [27.971 ns 28.345 ns 28.677 ns]
                        change: [-6.0848% -2.5856% +0.9298%] (p = 0.16 > 0.05)
                        No change in performance detected.
right shift/shr, U2048, large
                        time:   [189.17 ns 189.88 ns 190.73 ns]
                        change: [-1.0664% -0.2413% +0.6037%] (p = 0.58 > 0.05)
                        No change in performance detected.
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
right shift/shr, U2048, large, asm
                        time:   [28.857 ns 29.365 ns 29.835 ns]
                        change: [-2.7698% +0.8271% +4.5237%] (p = 0.66 > 0.05)
                        No change in performance detected.
Found 2 outliers among 100 measurements (2.00%)
  1 (1.00%) low mild
  1 (1.00%) high mild