fix: minimize `suppressWarnings` calls

From discussions in #319.

### `flatline_forecaster`

When I remove `suppressWarnings` [here](https://github.com/cmu-delphi/epipredict/blob/8a516e96bbd0d647291700c32da9d4c5182e9f13/R/flatline_forecaster.R#L72), that makes the tests [here](https://github.com/cmu-delphi/epipredict/blob/8a516e96bbd0d647291700c32da9d4c5182e9f13/tests/testthat/test-layer_residual_quantiles.R#L84-L89) fail. The warning is:

```
Detected an unexpected many-to-many relationship between `x` and `y`.
i Row 1 of `x` matches multiple rows in `y`.
i Row 1 of `y` matches multiple rows in `x`.
i If a many-to-many relationship is expected, set `relationship = "many-to-many"` to silence this warning.
```

### `cdc_baseline_forecaster`

If you remove the `suppressWarnings` there, you get the following in the tests

```
Using an external vector in selections was deprecated in tidyselect 1.1.0.
i Please use `all_of()` or `any_of()` instead.
# Was:
data %>% select(col_names)

# Now:
data %>% select(all_of(col_names))

Backtrace:
▆
1. ├─epipredict::cdc_baseline_forecaster(...) at test-snapshots.R:39:3
2. │ ├─... %>% ... at epipredict/R/cdc_baseline_forecaster.R:105:3
3. │ ├─stats::predict(wf, new_data = latest)
4. │ └─epipredict:::predict.epi_workflow(wf, new_data = latest)
5. │ ├─epipredict::apply_frosting(object, components, new_data, ...) at epipredict/R/epi_workflow.R:272:3
6. │ └─epipredict:::apply_frosting.epi_workflow(...) at epipredict/R/frosting.R:337:3
7. │ ├─epipredict::slather(la, components, workflow, new_data) at epipredict/R/frosting.R:403:7
8. │ └─epipredict:::slather.layer_unnest(la, components, workflow, new_data) at epipredict/R/layers.R:206:3
9. │ └─components$predictions %>% tidyr::unnest(col_names) at epipredict/R/layer_unnest.R:34:5
10. ├─dplyr::mutate(., target_date = forecast_date + ahead * args_list$data_frequency)
11. ├─dplyr::select(., -time_value)
12. ├─tibble::as_tibble(.)
13. ├─tidyr::unnest(., col_names)
14. ├─epiprocess:::unnest.epi_df(., col_names)
15. │ └─dplyr::dplyr_reconstruct(NextMethod(), data) at cmu-delphi-epiprocess-00c0a85/R/methods-epi_df.R:239:3
16. │ └─dplyr:::dplyr_new_data_frame(data)
17. │ ├─row.names %||% .row_names_info(x, type = 0L)
18. │ └─base::.row_names_info(x, type = 0L)
19. ├─base::NextMethod() at cmu-delphi-epiprocess-00c0a85/R/methods-epi_df.R:239:3
20. └─tidyr:::unnest.data.frame(., col_names)
21. └─tidyselect::eval_select(expr = enquo(cols), data = data, allow_rename = FALSE)
22. └─tidyselect:::eval_select_impl(...)
23. ├─tidyselect:::with_subscript_errors(...)
24. │ └─base::withCallingHandlers(...)
25. └─tidyselect:::vars_select_eval(...)
26. └─tidyselect:::walk_data_tree(expr, data_mask, context_mask)
27. └─tidyselect:::eval_sym(expr, data_mask, context_mask)
```

### `dist_quantiles`

Removing [this one](https://github.com/cmu-delphi/epipredict/blob/8a516e96bbd0d647291700c32da9d4c5182e9f13/R/dist_quantiles.R#L33) gave no warnings or errors in tests, but I'm not sure what this was intended to suppress, so I'm not comfortable removing that.