Adaptative URL filter to normalize URLs based on canonical tag

Such a filter could compare the parameters of a URL with the canonical tag found in the page (if any) and determine after a while which parameters can be safely removed in order to normalise the URL.

The aim is similar to the [clean-param extension of the robots protocol by Yandex](https://yandex.com/support/webmaster/controlling-robot/robots-txt.xml#clean-param) where sites can specify how URLs can be normalised.

TODO compare with [research.google.com/pubs/archive/35210.pdf]

Start by reviewing the issue proposal and comparing the clean-param extension with the linked Google research paper. Then locate the crawler's URL filtering and canonical-tag handling entry points; the work is complete when an adaptive filter can determine which URL parameters are safe to remove based on canonical tags.

Written by the indexing model from the issue text.