Fix World Age Problems

[willtebbutt]

I had introduced some fairly conservative world age checks in a previous PR, the purpose of which was to err on the side of caution, and insist that you only attempt to derive a rule based on the current world age. Unfortunately, this doesn't interact well with BenchmarkTools, because it typically progresses the world age, and you have to have constructed a rule by the time that a call to @benchmark is made.

This PR doesn't relax the world age assumption, but does make a point of updating the interpreter more regularly, to prevent it from getting behind.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Files with missing lines	Coverage Δ
src/interpreter/s2s_reverse_mode_ad.jl	93.16% <100.00%> (+0.01%)	⬆️

[github-actions]

Performance Ratio:
Ratio of time to compute gradient and time to compute function.
Warning: results are very approximate! See here for more context.

┌────────────────────────────┬────────┬─────────┬─────────────┬─────────┐
│                      Label │  Tapir │  Zygote │ ReverseDiff │  Enzyme │
│                     String │ String │  String │      String │  String │
├────────────────────────────┼────────┼─────────┼─────────────┼─────────┤
│                   sum_1000 │  108.0 │     0.9 │        5.41 │     1.8 │
│                  _sum_1000 │   7.96 │  1430.0 │        39.8 │   0.106 │
│               sum_sin_1000 │   2.43 │    1.59 │        10.7 │    1.01 │
│              _sum_sin_1000 │   3.09 │   355.0 │        17.0 │    1.51 │
│                   kron_sum │   62.8 │    10.6 │       201.0 │    9.99 │
│              kron_view_sum │   28.6 │    3.93 │        84.3 │    4.21 │
│      naive_map_sin_cos_exp │   3.18 │ missing │        10.1 │     2.8 │
│            map_sin_cos_exp │   4.68 │    1.76 │        7.52 │    3.58 │
│      broadcast_sin_cos_exp │    4.3 │    2.59 │        1.68 │    2.84 │
│                 simple_mlp │   8.11 │    3.11 │        13.6 │    3.16 │
│                     gp_lml │   17.1 │    6.42 │     missing │ missing │
│ turing_broadcast_benchmark │   7.24 │ missing │        36.3 │ missing │
│         large_single_block │   3.89 │  4040.0 │        30.5 │    2.23 │
└────────────────────────────┴────────┴─────────┴─────────────┴─────────┘