fix: treat .gitignore.in as a gitignore.in template, not a shell script

[kitsuyui]

Summary

.gitignore here is generated by the gitignore.in CLI, which parses .gitignore.in as a small template language and never executes it as a shell script.

The file still carried a vestigial shell wrapper from the era when it genuinely was an executable script:

#!/usr/bin/env bash
set -euo pipefail
cd "${0%/*}" && exec > .gitignore
gi() { curl -L -s https://www.gitignore.io/api/"$*"; }

This only kept working by coincidence: the parser ignores the wrapper lines while the bare gibo dump ... lines still match the gibo dump prefix. Editing the file as a shell script (e.g. wrapping the dumps in a block and indenting them) would silently disable every template, because the parser does not trim leading whitespace.

Changes

• .gitignore.in: reduce to the template form the tool expects (one directive per line, from column 0).
• .gitignore: regenerated with gitignore.in.

How to regenerate

$ gitignore.in

[github-actions]

🎉 Happy commit!

• dogfooding: PR fix: treat .gitignore.in as a gitignore.in template, not a shell script #842 always celebrates.
• dogfooding: 8e08818e07b68a6c3030370167e8f33d1fb48d19 always celebrates.

[github-actions]

Code Metrics Report

	main (6709cec)	#842 (b5be29c)	+/-
Coverage	99.2%	99.2%	0.0%
Code to Test Ratio	1:0.5	1:0.5	0.0
Test Execution Time	1s	1s	0s

Details

  |                     | main (6709cec) | #842 (b5be29c) | +/-  |
  |---------------------|----------------|----------------|------|
  | Coverage            |          99.2% |          99.2% | 0.0% |
  |   Files             |              5 |              5 |    0 |
  |   Lines             |            134 |            134 |    0 |
  |   Covered           |            133 |            133 |    0 |
  | Code to Test Ratio  |          1:0.5 |          1:0.5 |  0.0 |
  |   Code              |           1678 |           1678 |    0 |
  |   Test              |            955 |            955 |    0 |
  | Test Execution Time |             1s |             1s |   0s |

---

Reported by octocov