Simplify removing leading and trailing separators

[cottsay]

This change is adapted from ament/ament_package#152, authored by @robwoolley.

Previously, we checked if the leading or trailing character was a colon and then used a wildcard to remove it.

It is simpler to just remove a leading or trailing colon. This has the added benefit of only using shell built-in functions.

Requires tests that are added in #713.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 86.98%. Comparing base (3310754) to head (5eab722).
⚠️ Report is 1 commits behind head on cottsay/shell-tests.

Additional details and impacted files

@@                 Coverage Diff                  @@
##           cottsay/shell-tests     #712   +/-   ##
====================================================
  Coverage                86.98%   86.98%           
====================================================
  Files                       69       69           
  Lines                     4088     4088           
  Branches                   706      706           
====================================================
  Hits                      3556     3556           
  Misses                     421      421           
  Partials                   111      111           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[cottsay]

So this is sort of interesting, because I uncovered some issues in #713 that are solved by this PR, but not because of the lack of -c support for head or tail that originally motivated the change. It appears that on macOS, I have echo -n foo working as expected in zsh, but when I switch to sh, the /bin/echo executable that's used doesn't accept -n anymore so the conditional never fires.

It doesn't actually fail, it just doesn't remove the leading/trailing separators like it should.

Given that there's actually buggy behavior on a supported platform, I'm a little more motivated to get this pushed through and get #713 merged with some actual tests here.