This adds t1_name and t2_name parameters to DeepDiff so the t1/t2 labels in the tree view's string representation can be renamed, as suggested in #531.

Right now the DiffLevel repr hardcodes those labels:

>>> DeepDiff({'status': 'UP'}, {'status': 'DOWN'}, view='tree')
{'values_changed': [<root['status'] t1:"UP", t2:"DOWN">]}

With the new parameters you can give the two objects meaningful names:

>>> DeepDiff({'status': 'UP'}, {'status': 'DOWN'}, view='tree',
...          t1_name='before', t2_name='after')
{'values_changed': [<root['status'] before:"UP", after:"DOWN">]}

Notes on the implementation:

• The labels default to "t1"/"t2", so existing output is completely unchanged.
• They only affect the repr of tree view levels. The text view result and the comparison itself are untouched.
• The names are stored on the root DiffLevel and propagate to child levels through the existing create_deeper/copy machinery (same path verbose_level already uses), so nested nodes render with the custom labels too.
• Externally-supplied _parameters dicts that predate this change (and therefore don't include the keys) fall back to the defaults, so nothing breaks there.

Added tests in tests/test_model.py covering the custom labels, propagation to nested levels, the unchanged defaults, and that the text view is unaffected. Also updated the tree view docs, the parameter reference, CHANGELOG.md, and AUTHORS.md.

Closes #531