fix: prevent deadlock in Println/Printf after program shutdown

[Ricardo-M-L]

Fix Program.Println() and Printf() which sent directly to p.msgs without context check, unlike Send(). After shutdown these calls blocked forever, violating the documented safety contract.

[codecov]

Codecov Report

❌ Patch coverage is 0% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 57.72%. Comparing base (351d215) to head (f6a8be0).
⚠️ Report is 4 commits behind head on main.

Files with missing lines	Patch %	Lines
tea.go	0.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1779      +/-   ##
==========================================
+ Coverage   56.50%   57.72%   +1.22%     
==========================================
  Files          25       25              
  Lines        1315     1339      +24     
==========================================
+ Hits          743      773      +30     
+ Misses        482      476       -6     
  Partials       90       90              

☔ View full report in Codecov by Harness.
📢 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.

[Ricardo-M-L]

Friendly ping - this PR is small and ready for review whenever there's bandwidth.

What it fixes: Program.Println / Printf send via p.msgs <- directly, which is an unbuffered channel send. Once the program has exited, the eventLoop goroutine that drains p.msgs is gone, so a late Println blocks forever and leaks the calling goroutine (deadlock). The fix routes through p.Send(...), which is already guarded against a closed/stopped program.

Scope: 1 hunk in tea.go. The tutorials/go.mod / go.sum touchups are from a go mod tidy while building the examples - happy to drop them if you'd rather keep this PR to a single file.

Overlap note: this branch also carries the recoverFromGoPanic cleanup fix (p.cancel() -> p.shutdown(true)) which is filed separately as PR 1770. If you'd prefer one PR per fix, I can split that hunk out so this one is only the Println/Printf change.