[MRG] Update CHANGELOG for the Gensim 4.1.0 release

[piskvorky]

Fixes #3199.

@mpenkov I also linked from CHANGELOG to the new EnsambleLDA tutorial which isn't "live" yet: https://radimrehurek.com/gensim/auto_examples/tutorials/run_ensemblelda.html

Once we release & rebuild & upload the docs, the link should work. But let's verify.

[mpenkov]

@piskvorky I added additional info about backwards incompatibilities and fixed some typos, please have a look.

[piskvorky]

Thanks for the fixes, looks good! Sorry the PR took me so long, I was sick last week.

[piskvorky]

FYI, running the generate_changelog.py script failed with some 403 API error:

I contacted the Github support, this is their answer:

Thanks for writing in to GitHub Support!

I was able to find in our logs some of the instances of the request made to that endpoint( https://api.github.com/repos/RaRe-Technologies/gensim...).

You're hitting the abuse rate limits and are different from our primary rate limits which are documented here:

https://docs.github.com/free-pro-team@latest/rest/overview/resources-in-the-rest-api#rate-limiting

The abuse rate limits are primarily here to ensure the API's availability by limiting spikes in requests, especially expensive requests. This normally happens when users make lots of requests in a short period of time, especially requests that consume a lot of resources, and also when making requests concurrently.

There's documentation on the abuse rate limits here along with advice on avoiding them and what to do if you do hit them:

https://docs.github.com/free-pro-team@latest/rest/guides/best-practices-for-integrators#dealing-with-abuse-rate-limits

The abuse rate limits are necessary and ensure the stability of the platform for all customers; we're not able to disable or increase them. We recommend checking out that guide to make sure you're able to handle these responses and let us know if you have any other questions about this!

Looks like the solution will be to simply wait 1 sec between the Github API requests.