Release v3.3.0

[RomiconEZ]

Update version to 3.3.0

[Copilot]

Pull Request Overview

A release bump to version 3.3.0 that adds a safe default for num_attempts and updates tests and examples accordingly

• Introduce _DEFAULT_NUM_ATTEMPTS and patch logic to ensure num_attempts falls back to 3
• Bump __version__, docs, and bumpversion config to 3.3.0
• Update tests and example notebooks to reflect the new default num_attempts

Reviewed Changes

Copilot reviewed 12 out of 12 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
tests/openai_client_lm_studio_test.py	Expand test cases to cover multiple presets
src/llamator/utils/attack_params.py	Add _DEFAULT_NUM_ATTEMPTS and patch num_attempts
src/llamator/version.py	Bump package version to 3.3.0
docs/howtos.md & README.md	Update installation instructions to 3.3.0
.bumpversion.cfg	Update bumpversion current_version to 3.3.0
examples/*.ipynb	Refresh notebook execution metadata and outputs

Comments suppressed due to low confidence (2)

examples/llamator-whatsapp.ipynb:43

• [nitpick] These notebooks include updated outputs and execution metadata, causing very large diffs. Consider clearing cell outputs and execution counts before committing to reduce noise.

   "execution_count": 1

tests/openai_client_lm_studio_test.py:36

• The test for aim_jailbreak only asserts num_attempts, but the implementation likely also returns default parameters like custom_dataset or language. Include all expected default keys in the assertion to keep the test accurate.

        ("aim_jailbreak", {"num_attempts": 3}),