CI: new action to run visual tests

test: make the visual test standalone (launch its own app instance)

test: added expected args to the nooop guard for multi-run mock

CI: longer timeout for slow test (my laptop faster than ubuntu-latest runner)

docs: mkdocs needs modules (adding __init__.py)

CI: removed this branch from triggers, testing / fixes done

docs: added some notes on running the visual tests

docs: repaired docs following restructuring done in history

CI: reorganised tests so can run most without selenium installation

CI: filter out visual tests; temp add this branch to verify

test: updated markers for cleaner filtering

chore: adding separate requirements file for selenium testing

test: random images with relevant metadata

test: visual test of presented content, persistent to tab switching

chore: added custom marker to slow (>20s) test, updated CI to skip

test: basic measure of timing for each phase

test: fixed error in indexing for the observation overrides

test: end-to-end test of upload -> inference

test: debugging the mock / side-effect usage

test: reorganised to be more modular

fix: all usage of session_state.files explicitly a list now

test: implemented demo/snippet using app code for sidebar, plus test

test: now using fixture that loads real data in the file_upload

test: new test for input validation, and fix typo in args

chore: adding a property and tidy up fstring warnings

test: first steps with AppTest and a mock multi-file file_uploader

test: for setters and for to_dict method of InputObservation

test: add test for equality of InputObservation objects

test: InputObservation with valid and invalid inputs

test: mock of FileUploader and tiny test validating it

feat: extra CI action that adds coverage reports to PR conversation

chore: tidy up main

fix: retired use of tab_log in session state

chore: moved all session state init to relevant modules

fix: typehints compatible with py3.9

doc: added docstrings to new functions

fix: remove highlighting for completion, not valid before v1.41.0

chore: cleanup variable names and add docstrings

chore: moved debug function to relevant module, and docstring

chore: clean up function naming

test: updated test to match interface; solved other xfail tests

feat: extended InputObservation to contain species/prediction info

chore: tidy up of workflow and debug clutter

fix: InputObservation uses date, time, and raw_image_datetime

feat: push observations functions for multi-file handling

fix: instantiating InputObservation with new call signature

fix: public_observations now a dict, for multi-image handling

fix: removed time from InputObservation, renamed date for clarity

chore: removing duplicate file (right one is in classifier subdir)

feat: get coordinates from file, populate input boxes