fix(plugin-dev): validator scripts abort on first finding due to set -e

[wellkilo]

Problem

Three validator scripts in plugin-dev use set -euo pipefail:

• plugins/plugin-dev/skills/agent-development/scripts/validate-agent.sh
• plugins/plugin-dev/skills/hook-development/scripts/hook-linter.sh
• plugins/plugin-dev/skills/hook-development/scripts/validate-hook-schema.sh

These are "check each item → accumulate errors/warnings → print a summary"
validators. But several commands they rely on legitimately return non-zero
during normal operation, and with -e any of them aborts the whole script
at the first finding — before the summary or correct exit code is reached:

• ((counter++)) returns exit code 1 when the counter goes from 0 → 1
  (post-increment evaluates to the old value 0, which is arithmetically false).
• grep '^description:' returns non-zero when an optional field is absent.
• jq returns non-zero when indexing a malformed structure.

Result: the validators silently fail exactly when the input has problems —
the case they exist to handle.

Repro

$ bash validate-agent.sh some-agent-with-errors.md
# stops at the first ❌, no summary, output truncated

Fix

Change set -euo pipefail → set -uo pipefail in all three scripts.
-u and pipefail are kept. All fatal conditions (missing file, missing
args, malformed frontmatter) already use explicit exit 1, so error handling
behavior is unchanged.

Verification

• Malformed inputs now run all checks and print the correct summary + exit code.
• Valid inputs still exit 0 with "✅ All checks passed!".
• bash -n passes on all three files.

[stevei101]

Approved by Antigravity AI pair programmer after verifying CI checks pass.

[wellkilo]

@ashwin-ant Hi, This PR fixes the early exit issue in plugin-dev validator scripts when using -e.
The changes have been reviewed and approved by @stevei101, but I still need a maintainer's approval for the workflow and merge to proceed.
Could someone please take a look when you get a chance?
Thanks a lot!