CompatHelper: bump compat for GPUArraysCore to 0.2, (keep existing compat)

[github-actions]

This pull request changes the compat entry for the GPUArraysCore package from 0.1 to 0.1, 0.2.
This keeps the compat entries for earlier versions.

Note: I have not tested your package with this new compat entry.
It is your responsibility to make sure that your package tests pass before you merge this pull request.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

[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 │ Mooncake │  Zygote │ ReverseDiff │  Enzyme │
│                     String │   String │  String │      String │  String │
├────────────────────────────┼──────────┼─────────┼─────────────┼─────────┤
│                   sum_1000 │     83.0 │     1.1 │        5.61 │    8.11 │
│                  _sum_1000 │     6.97 │  1410.0 │        34.0 │    1.09 │
│               sum_sin_1000 │     2.26 │     1.7 │        10.8 │    1.99 │
│              _sum_sin_1000 │     2.68 │   256.0 │        13.5 │    2.45 │
│                   kron_sum │     66.5 │    3.59 │       208.0 │    7.88 │
│              kron_view_sum │     20.8 │    3.35 │        87.1 │    34.0 │
│      naive_map_sin_cos_exp │     2.53 │ missing │        7.17 │    2.34 │
│            map_sin_cos_exp │     2.75 │    1.54 │        6.14 │    2.92 │
│      broadcast_sin_cos_exp │     2.56 │    2.26 │        1.46 │    2.29 │
│                 simple_mlp │     7.61 │    2.91 │        11.5 │     3.6 │
│                     gp_lml │     12.8 │    10.6 │     missing │    9.62 │
│ turing_broadcast_benchmark │     3.18 │ missing │        25.8 │ missing │
│         large_single_block │     4.42 │  4780.0 │        31.2 │    2.18 │
└────────────────────────────┴──────────┴─────────┴─────────────┴─────────┘

[willtebbutt]

This actually looks fine -- NNlib doesn't appear to be compatible with 0.2.0 yet, but it appears to be the case that everything else currently works. I'm going to merge this.