update to latest zig-lsp-codegen with new LSP Message type/parser

[Techatrix]

This includes a new parser for the LSP json messages. Its up to 20% faster and uses half as much memory.
Binary size of ZLS goes up by about 1-3% but that is a trade off I am willing to take.

[nix-shell:~/repos/zig-lsp-codegen/zig-out/bin]$ poop ./bench-old-debug ./bench-new-debug
Benchmark 1 (3 runs): ./bench-old-debug
  measurement          mean ± σ            min … max           outliers         delta
  wall_time          17.9s  ±  113ms    17.9s  … 18.1s           0 ( 0%)        0%
  peak_rss           30.9MB ± 14.4KB    30.9MB … 30.9MB          0 ( 0%)        0%
  cpu_cycles         62.7G  ±  220M     62.6G  … 63.0G           0 ( 0%)        0%
  instructions        122G  ± 6.10K      122G  …  122G           0 ( 0%)        0%
  cache_references   77.6M  ± 9.07M     70.0M  … 87.7M           0 ( 0%)        0%
  cache_misses       3.99M  ±  830K     3.49M  … 4.95M           0 ( 0%)        0%
  branch_misses      92.0M  ±  164K     91.8M  … 92.2M           0 ( 0%)        0%
Benchmark 2 (3 runs): ./bench-new-debug
  measurement          mean ± σ            min … max           outliers         delta
  wall_time          15.4s  ±  314ms    15.0s  … 15.6s           0 ( 0%)        ⚡- 14.3% ±  3.0%
  peak_rss           15.9MB ± 2.36KB    15.9MB … 15.9MB          0 ( 0%)        ⚡- 48.4% ±  0.1%
  cpu_cycles         55.0G  ±  795M     54.1G  … 55.6G           0 ( 0%)        ⚡- 12.4% ±  2.1%
  instructions        105G  ± 8.31K      105G  …  105G           0 ( 0%)        ⚡- 13.9% ±  0.0%
  cache_references   97.8M  ± 9.71M     86.6M  …  104M           0 ( 0%)          + 26.1% ± 27.4%
  cache_misses       4.50M  ±  582K     3.84M  … 4.92M           0 ( 0%)          + 12.8% ± 40.7%
  branch_misses      75.6M  ±  849K     74.6M  … 76.1M           0 ( 0%)        ⚡- 17.8% ±  1.5%

[nix-shell:~/repos/zig-lsp-codegen/zig-out/bin]$ poop -d 20000 ./bench-old-safe ./bench-new-safe
Benchmark 1 (6 runs): ./bench-old-safe
  measurement          mean ± σ            min … max           outliers         delta
  wall_time          3.62s  ± 26.3ms    3.59s  … 3.65s           0 ( 0%)        0%
  peak_rss           29.9MB ± 62.8KB    29.8MB … 30.0MB          0 ( 0%)        0%
  cpu_cycles         10.4G  ± 49.6M     10.4G  … 10.5G           0 ( 0%)        0%
  instructions       24.0G  ± 18.3M     24.0G  … 24.0G           0 ( 0%)        0%
  cache_references   16.0M  ±  398K     15.6M  … 16.6M           0 ( 0%)        0%
  cache_misses       1.89M  ± 82.5K     1.77M  … 2.02M           0 ( 0%)        0%
  branch_misses       116M  ± 92.9K      116M  …  116M           0 ( 0%)        0%
Benchmark 2 (8 runs): ./bench-new-safe
  measurement          mean ± σ            min … max           outliers         delta
  wall_time          2.81s  ± 21.7ms    2.79s  … 2.84s           0 ( 0%)        ⚡- 22.3% ±  0.8%
  peak_rss           14.6MB ± 13.7KB    14.6MB … 14.6MB          3 (38%)        ⚡- 51.1% ±  0.2%
  cpu_cycles         9.27G  ± 55.4M     9.20G  … 9.35G           0 ( 0%)        ⚡- 11.0% ±  0.6%
  instructions       21.7G  ±  133      21.7G  … 21.7G           0 ( 0%)        ⚡-  9.5% ±  0.1%
  cache_references   8.15M  ±  406K     7.57M  … 8.82M           0 ( 0%)        ⚡- 49.1% ±  3.0%
  cache_misses       1.18M  ± 23.9K     1.14M  … 1.23M           0 ( 0%)        ⚡- 37.4% ±  3.5%
  branch_misses       108M  ±  351K      107M  …  108M           1 (13%)        ⚡-  7.3% ±  0.3%

[nix-shell:~/repos/zig-lsp-codegen/zig-out/bin]$ poop -d 20000 ./bench-old-fast ./bench-new-fast
Benchmark 1 (6 runs): ./bench-old-fast
  measurement          mean ± σ            min … max           outliers         delta
  wall_time          3.37s  ± 19.0ms    3.35s  … 3.39s           0 ( 0%)        0%
  peak_rss           28.2MB ± 3.34KB    28.2MB … 28.2MB          0 ( 0%)        0%
  cpu_cycles         9.67G  ± 53.2M     9.61G  … 9.75G           0 ( 0%)        0%
  instructions       20.1G  ± 1.30K     20.1G  … 20.1G           0 ( 0%)        0%
  cache_references   10.1M  ±  447K     9.63M  … 10.9M           0 ( 0%)        0%
  cache_misses       1.86M  ±  608K     1.56M  … 3.10M           0 ( 0%)        0%
  branch_misses       133M  ±  415K      133M  …  134M           0 ( 0%)        0%
Benchmark 2 (8 runs): ./bench-new-fast
  measurement          mean ± σ            min … max           outliers         delta
  wall_time          2.66s  ± 12.5ms    2.65s  … 2.68s           0 ( 0%)        ⚡- 21.0% ±  0.5%
  peak_rss           14.1MB ± 3.10KB    14.1MB … 14.1MB          0 ( 0%)        ⚡- 50.1% ±  0.0%
  cpu_cycles         8.77G  ± 42.6M     8.72G  … 8.85G           0 ( 0%)        ⚡-  9.3% ±  0.6%
  instructions       18.8G  ± 95.6      18.8G  … 18.8G           0 ( 0%)        ⚡-  6.5% ±  0.0%
  cache_references   5.97M  ±  117K     5.79M  … 6.13M           2 (25%)        ⚡- 41.0% ±  3.5%
  cache_misses       1.06M  ± 45.9K      986K  … 1.13M           0 ( 0%)        ⚡- 43.1% ± 24.9%
  branch_misses       124M  ± 54.1K      124M  …  124M           0 ( 0%)        ⚡-  7.0% ±  0.2%

[nix-shell:~/repos/zig-lsp-codegen/zig-out/bin]$ poop -d 20000 ./bench-old-small ./bench-new-small
Benchmark 1 (5 runs): ./bench-old-small
  measurement          mean ± σ            min … max           outliers         delta
  wall_time          4.41s  ± 35.8ms    4.37s  … 4.46s           0 ( 0%)        0%
  peak_rss           28.0MB ± 3.43KB    28.0MB … 28.0MB          0 ( 0%)        0%
  cpu_cycles         13.5G  ± 59.0M     13.4G  … 13.6G           0 ( 0%)        0%
  instructions       31.9G  ± 1.01K     31.9G  … 31.9G           0 ( 0%)        0%
  cache_references   8.61M  ±  531K     8.03M  … 9.26M           0 ( 0%)        0%
  cache_misses       1.45M  ± 54.2K     1.39M  … 1.51M           0 ( 0%)        0%
  branch_misses       109M  ±  135K      109M  …  110M           0 ( 0%)        0%
Benchmark 2 (5 runs): ./bench-new-small
  measurement          mean ± σ            min … max           outliers         delta
  wall_time          4.45s  ± 8.87ms    4.44s  … 4.46s           0 ( 0%)          +  1.0% ±  0.9%
  peak_rss           14.0MB ± 17.0KB    13.9MB … 14.0MB          0 ( 0%)        ⚡- 50.1% ±  0.1%
  cpu_cycles         15.3G  ± 45.3M     15.3G  … 15.4G           0 ( 0%)        💩+ 13.8% ±  0.6%
  instructions       43.5G  ± 1.61K     43.5G  … 43.5G           0 ( 0%)        💩+ 36.4% ±  0.0%
  cache_references   6.33M  ±  472K     5.62M  … 6.94M           0 ( 0%)        ⚡- 26.5% ±  8.5%
  cache_misses       1.01M  ± 47.9K      949K  … 1.07M           0 ( 0%)        ⚡- 30.5% ±  5.2%
  branch_misses      90.5M  ± 89.7K     90.4M  … 90.6M           0 ( 0%)        ⚡- 17.3% ±  0.2%

I benchmarked it with messages that got collected from ZLS while I was writing the benchmarking code.