Categorise short report repos and clarify publication policy

[bloodearnest]

Our policy is publication of all repos after 12 months.

However, we have a class of repo that is a bit different - so called "short reports".

A lot of these still use github releases, and some have specific release/IG approval processes.

We need to make a list of all these repos, collaborating with researchers:

• tag them with the short-report topic
• determine if they are still active and current lead researcher/owner.
• determine if they are still using github release process
• determine what the current IG approval process is

Using this list, we need to then:

• decide if the 12 month policy applies to them (policy decision)
• decide if we can require short reports to be public from the get go (policy decision).
• plan to move the active ones still using github releases to using job-server releases.
• make them public once thats done.

[bloodearnest]

All short report repos have had the short-report topic/tag added.

osrelease logs showed the following for github releases

2022-01-12  SRO-Measures                    LouisFisher
2022-01-14  mechanical-valve-anticoag       LouisFisher
2022-01-17  long-covid-descriptive-research Robinpark
2022-02-03  SGTF-omi-research               DanielGrint
2022-02-04  principle-inhaler-uptake        jonmassey
2022-02-18  immunosuppressant-meds-research LouisFisher
2022-03-15  sro-gastro                      MillieGreen
2022-03-30  covid-vaccine-preliminary-study lisahopcroft
2022-04-01  comparative-ve-research         WillHulme
2022-04-22  hh-classification-research      Kevinwing
2022-05-13  covid-vaccine-neuro-research    Annaschultze

This means no one has been using github releases via osrelease, so we can go ahead an disable that.

However, even though they are being released to job-server, in many cases they are also being manually pushed to the github repo in order for reports to consume them. These need migrating to use the job-server release directly, have filed opensafely-core/reports#341 to track that.

[bloodearnest]

Regards review, it seems that pdfs are sent to reviewers via email, so there's no additional work needed here.