test(parser): snapsot testing with insta

[39555]

This draft includes an example of testing with insta. It requires serde support because Debug snapshots are unmanageably wide:

compared to yaml snapshots:

Usage

• pass the parsed ast into insta::assert_yaml_snapshot
• run with cargo insta review
• accept the snapshot
• rerun with cargo test
• if a newly created snapshot differs the test fails

Cons

One inconvenience is that multiline fields in snap files are not formatted which requires always using cargo insta tool to pretty printing snapshots. This might related to mitsuhiko/insta#372 I heven't found a good way to store the text input along with the ast snapshot.

---
source: brush-parser/src/parser.rs
description: "\n#!/usr/bin/env bash\n\nfor f in A B C; do\n\n    # sdfsdf\n    echo \"${f@L}\" >&2\n\n   done\n\n"
---
...

[github-actions]

Performance Benchmark Report

Benchmark name	Baseline (μs)	Test/PR (μs)	Delta (μs)	Delta %
expand_one_string	3.46 μs	3.46 μs	0.00 μs	⚪ Unchanged
instantiate_shell	61.00 μs	60.23 μs	-0.77 μs	⚪ Unchanged
instantiate_shell_with_init_scripts	30536.96 μs	31015.17 μs	478.21 μs	🟠 +1.57%
parse_bash_completion	2799.21 μs	2822.63 μs	23.42 μs	🟠 +0.84%
parse_sample_script	4.25 μs	4.22 μs	-0.03 μs	⚪ Unchanged
run_echo_builtin_command	90.72 μs	91.47 μs	0.75 μs	⚪ Unchanged
run_one_builtin_command	108.64 μs	110.67 μs	2.03 μs	⚪ Unchanged
run_one_external_command	1913.96 μs	1974.41 μs	60.45 μs	🟠 +3.16%
run_one_external_command_directly	1013.32 μs	1017.33 μs	4.01 μs	⚪ Unchanged

Code Coverage Report: Only Changed Files listed

Package	Base Coverage	New Coverage	Difference
brush-core/src/jobs.rs	🔴 42.42%	🔴 37.23%	🔴 -5.19%
brush-parser/src/parser.rs	🟢 99.14%	🟢 99.16%	🟢 0.02%
Overall Coverage	🟢 77.84%	🟢 77.78%	🔴 -0.06%

Minimum allowed coverage is 70%, this run produced 77.78%