BOT: Fix #780: Add verbose argument to as_forecast_*() constructors

[nikosbosse]

Summary

• Adds a verbose = TRUE parameter to all 7 as_forecast_<type>.default() methods
• Passes verbose through to assert_forecast() to suppress messages/warnings when verbose = FALSE
• Gates the quantile rounding cli_warn() in as_forecast_quantile.default() behind the verbose flag
• Adds @inheritParams assert_forecast roxygen tags for documentation

Fixes #780

Root cause

The verbose parameter existed on assert_forecast() and assert_forecast_generic() (added in PR #778), but was never threaded through to the as_forecast_<type>() constructors. Users were forced to wrap calls in suppressMessages()/suppressWarnings() to silence informational output.

What the fix does

Each .default() method now accepts verbose = TRUE and forwards it to assert_forecast(data, verbose = verbose). The quantile constructor additionally gates its own cli_warn() about rounding issues behind if (verbose). Default behavior (verbose = TRUE) is unchanged.

Test coverage added

• as_forecast_quantile() verbose suppresses NA messages, rounding warnings, and unequal-quantile warnings
• as_forecast_quantile() verbose = FALSE does not suppress errors
• as_forecast_quantile() default verbose = TRUE preserves messages
• as_forecast_sample() verbose suppresses NA messages
• as_forecast_binary() verbose suppresses NA messages
• as_forecast_point() verbose suppresses NA messages
• as_forecast_nominal() verbose works correctly

Test plan

• All new tests pass
• Full test suite passes (692 tests, 0 failures)
• R CMD check: 0 errors, 0 warnings, 2 pre-existing notes

🤖 Generated with Claude Code

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.83%. Comparing base (ac0c01a) to head (f511dc9).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1089   +/-   ##
=======================================
  Coverage   97.83%   97.83%           
=======================================
  Files          35       35           
  Lines        1845     1846    +1     
=======================================
+ Hits         1805     1806    +1     
  Misses         40       40           

☔ 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.

[nikosbosse]

CLAUDE: Approve. Clean, well-scoped implementation that correctly threads the existing verbose parameter from assert_forecast() through all 7 as_forecast_.default() constructors. The change is minimal, follows existing patterns, and preserves default behavior. Quantile rounding cli_warn() is properly gated while the rounding behavior itself still executes. Tests are thorough for quantile (6 tests) and adequate for other types. Minor gaps: no tests for ordinal/multivariate_sample types (low risk, identical code pattern) and the nominal test uses na.omit() so doesn't exercise actual suppression. The optional cleanup of existing suppressMessages/suppressWarnings wrappers was reasonably deferred to a follow-up.