Add docs

[MechaCritter]

Related Issues

Adds missing technical documentation for future developers.

Proposed Changes:

Related Issues

• N/A (documentation-only addition)

Proposed Changes:

This adds a docs/ directory with technical reference documentation for the entire
pyvisim package. There were no existing prose docs - just the source code and
docstrings - so this is a from-scratch addition.

What's included:

• docs/overview.md: a developer overview covering package layout, the core
  extract → aggregate → normalize pipeline, the two abstract interfaces
  (SimilarityMetric, FeatureExtractorBase), and design decisions worth knowing
  before touching the code (NumPy-only images, the pluggable similarity function, enum
  weights).
• docs/encoders/: reference pages for VLADEncoder, FisherVectorEncoder,
  Pipeline, ImageEncoderBase, and the pretrained weights enums. Includes a
  VLAD-vs-Fisher Vector comparison and an explanation of the automatic PCA pairing.
• docs/features/: reference pages for SIFT, RootSIFT, DeepConvFeature, and
  Lambda, including the Hellinger normalization rationale for RootSIFT and how
  spatial encoding affects output_dim in DeepConvFeature.
• docs/dataset/: documents OxfordFlowerDataset and calls out the deliberate
  train/test split swap (original test → "train", original train → "test").
• docs/neural_networks/: placeholder noting the Siamese network is planned but not
  yet implemented.

How did you test it?

Documentation only — no code changes. I read every linked source file to verify the
descriptions match the actual implementation.

Notes for the reviewer

The OxfordFlowerDataset train/test swap is documented explicitly because it's
non-obvious and will bite anyone who compares retrieval numbers against papers using
the standard split. Worth double-checking that the description in
docs/dataset/oxford_flower_dataset.md matches what's actually in the source.

Checklist

• I have read the contributors guidelines.
• I have updated the related issue with new insights and changes.
• I have added unit tests and updated the docstrings.
• I've used one of the conventional commit types for my PR title: docs: add technical reference documentation for pyvisim
• I have documented my code.
• When applicable: I have reviewed the AI-generated code sections, according to contributors guidelines.
• I have run pre-commit hooks and fixed any issue.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.