mark dry run mode in poetry output

Issue Kind

Change in current behaviour

Description

Currently, there is no difference in output between poetry install --dry-run and poetry install.

$ poetry install --dry-run
Installing dependencies from lock file

Package operations: 4 installs, 0 updates, 0 removals

  - Installing mdurl (0.1.2)
  - Installing pygments (2.18.0)
  - Installing markdown-it-py (3.0.0)
  - Installing rich (13.9.4)

Installing the current project: my-package (0.1.0)

$ poetry install
Installing dependencies from lock file

Package operations: 4 installs, 0 updates, 0 removals

  - Installing mdurl (0.1.2)
  - Installing pygments (2.18.0)
  - Installing markdown-it-py (3.0.0)
  - Installing rich (13.9.4)

Installing the current project: my-package (0.1.0)

Impact

This may be confusing, since observing only the output of poetry command, there is no way of telling if it was actually executed, or only simulated.

Workarounds

You need to observe both the command used (to check for dry-run flag) and the output (to see an actual output) in order to check if for example a package was installed. This makes scripting poetry output more challenging.

Start by tracing the output produced by poetry install and how the --dry-run option is handled. Compare the dry-run and normal execution paths, then add or update coverage for their output. Done means the output clearly indicates when installation is simulated, without requiring users to inspect the command arguments.