_keep_num() was incorrectly detecting overlaps

Otherwise features tend to cluster

As with VAFE, update LoudFE to expand to min duration when keeping n features

see: 72fc6b563b

The problem that I had overlooked until now (!!) due to only using small
exemplar videos for testing was that VideoActivityFeatureExtractor keeps
a fraction of small (0.5s) Features. This is not so problematic for a short
exemplar video, but ends up witha  lot of jumpy Features in an actual source
video.

Fix approach:

Like LoudnessFE, keep a specific number and expand the duration (which I think I
will do for LoudnessFE too), dropping any that would be consumed in that range

add LoudAudioFeatureExtractor._num_features (set in init, default: 5)

add _keep_num -- takes list of loudnesses and keeps top N (default: 5)

Check for missing input files not needed as object cannot be initialised without them

Calls pulled out relate to setup and working of Whisper:
 - _whispermodel()
 - _batched_inference_pipeline()
 - _transcribe()

Defaults defined: model, device, compute type, beamsize, batchsize, pipeline type

Tests:
 - basic init
 - init with no media
 - run() with no words (early exit 0 Features)
 - run() with mocked transcribe

NOTE: these are unit tests and do not exercise Whisper

BREAKING CHANGE: no words to WFE are no longer an error, they raise a notice

WordFeatureExtractor is not fast- even the import is slow. However, it processes
files and returns Features corresponding to matched words.

WhisperFE will be slightly different to other FEs in that there is/are specific
target words to be searched for. Not specifying these could be an error (this
commit specifies this as such) but a better approach may be to downgrade that to
a (logging) notice, and simply match nothing / early exit.

To help functional testing, LaughFE's internal adjustment times are exposed.

Recap: when a laugh is detected by LaughFE, the time of the laugh itself it not
used directly; instead, the resulting Feature has some time prepended to try to
capture the thing that caused the laugh.

When functional testing the FEs we set up specially-crafted videos with features
at known points, so to make sure the LaughterFE is tested correctly we adjust
the tests by the amount of time the FE adjusts by so that it properly tests the
intended behaviour.

@see:
 - feature_extractors.py::LaughterFeatureExtractor

This was done so the collection of loudnesses from an audio file could be mocked
in testing, but improves readability

TODO: review number of params and consider further refactoring

Runs a command returns stout as string, defined to aid testing

facilitates test mocking

Parses JSON, eg that output by the pipeline or defined manually and
produces Features that can be (re-)used with the pipeline

Take the mean of non-overlapping windows of scores

    Input:  list of tuples in the format (time, score)
    Output: list of tuples in the format (time, mean_score)
            (reduced set)

Drops lowest n% (default:33%) of scdet scores, since it scores every
frame

Python being what it is, this could be a single line in another method
but pulling it out into another function:

 - makes explicit what we are doing and lets us document why
 - makes for easier testing

Uses ffmpeg's scdet under the hood apropos of recommendation on
ffmpeg-user ML [1]

[1]: https://ffmpeg.org/pipermail/ffmpeg-user/2024-August/058535.html

Uses pyloudnorm under the hood to determine the loudness of the supplied
media file (handles videos transparently)

TBC: some sort of limiter on the number produced

TBC

TBC

Simple feature extractor that picks random times, used to exercise other
parts of the pipeline

This adds functionality for getting laughter by using jrgillick's
laughter detection library

NB python expects all of the feature extractor's dependencies to be
available; perhaps in future we can do something even fancier like
activating another python env

[retroactive commit]

for jrgillick/laughter-detection to slot into

or at least what passes for an interface in python