Improve diff lines count efficiency

Prepares us for upgrade to Rails 5.2

The `#reload` makes to load all objects into memory,
and the main purpose of `#reload` is to drop the association cache.

The `#reset` seems to solve exactly that case.

Since external diffs are likely to be a bit slower than in-database
ones, add a mode that makes diffs external after they've been obsoleted
by events. This should strike a balance between performance and disk
space.

A background cron drives the majority of migrations, since diffs become
outdated through user actions.

If a diff happened to include a single null byte anywhere, insertion
into the database would fail with an Error 500 since the column is text
and not a byte array. To fix this, we mark the diff as binary if we
detect a single null byte and Base64-encode it.

Closes https://gitlab.com/gitlab-org/gitlab-ce/issues/57710

When a branch is deleted, the push commit IDs is an empty array.
The previous change would attempt to call `[].exists?`, which is
invalid. Fix this by returning `MergeRequestDiffCommit.none` instead.

Closes https://gitlab.com/gitlab-org/gitlab-ce/issues/53853

Fetch database columns directly if available

Previously for a given merge request, we would:

1. Create the array of commit SHAs involved in the push (A)
2. Request all merge request commits and map the SHA (B)
3. Reload the diff if there were any common commits between A and B

We can avoid additional database querying and overhead by
checking with the database whether the merge request contains any
of the commit SHAs.

Relates to https://gitlab.com/gitlab-org/gitlab-ce/issues/53213

This whitelists all existing offenses for the various CodeReuse cops, of
which most are triggered by the CodeReuse/ActiveRecord cop.

Partially addresses #47424.

* processes any pending records which are not migrated yet
* bumps import_export version because of new commits_count attribute
* removes commits_count fallback method

This reverts commit 0b7d1085.

* processes any pending records which are not migrated yet
* bumps import_export version because of new commits_count attribute
* removes commits_count fallback method

This reduces the number of RefExists RPC call.

For each MR diff an extra 'SELECT COUNT()' is executed
to get number of commits for the diff. Overall time to get counts for
all MR diffs may be quite expensive. To speed up loading of MR info,
information about number of commits is stored in a MR diff's extra column.

Closes #38068

The Gitaly CommitService is being hammered by n + 1 calls, mostly when
finding commits. This leads to this gRPC being turned of on production:
https://gitlab.com/gitlab-org/gitaly/issues/514#note_48991378

Hunting down where it came from, most of them were due to
MergeRequest#show. To prove this, I set a script to request the
MergeRequest#show page 50 times. The GDK was being scraped by
Prometheus, where we have metrics on controller#action and their Gitaly
calls performed. On both occations I've restarted the full GDK so all
caches had to be rebuild.

Current master, 806a68a8, needed 435 requests
After this commit, 154 requests

The st_commits and st_diffs columns on merge_request_diffs historically held the
YAML-serialised data for a merge request diff, in a variety of formats.

Since 9.5, these have been migrated in the background to two new tables:
merge_request_diff_commits and merge_request_diff_files. That has the advantage
that we can actually query the data (for instance, to find out how many commits
we've stored), and that it can't be in a variety of formats, but must match the
new schema.

This is the final step of that journey, where we drop those columns and remove
all references to them. This is a breaking change to the importer, because we
can no longer import diffs created in the old format, and we cannot guarantee
the export will be in the new format unless it was generated after this commit.

Compared to the merge_request_diff association:

1. It's simpler to query. The query uses a foreign key to the
   merge_request_diffs table, so no ordering is necessary.
2. It's faster for preloading. The merge_request_diff association has to load
   every diff for the MRs in the set, then discard all but the most recent for
   each. This association means that Rails can just query for N diffs from N
   MRs.
3. It's more complicated to update. This is a bidirectional foreign key, so we
   need to update two tables when adding a diff record. This also means we need
   to handle this as a special case when importing a GitLab project.

There is some juggling with this association in the merge request model:

* `MergeRequest#latest_merge_request_diff` is _always_ the latest diff.
* `MergeRequest#merge_request_diff` reuses
  `MergeRequest#latest_merge_request_diff` unless:
    * Arguments are passed. These are typically to force-reload the association.
    * It doesn't exist. That means we might be trying to implicitly create a
      diff. This only seems to happen in specs.
    * The association is already loaded. This is important for the reasons
      explained in the comment, which I'll reiterate here: if we a) load a
      non-latest diff, then b) get its `merge_request`, then c) get that MR's
      `merge_request_diff`, we should get the diff we loaded in c), even though
      that's not the latest diff.

Basically, `MergeRequest#merge_request_diff` is the latest diff in most cases,
but not quite all.

This adds an optimised way of getting the latest pipeline status for a
list of Commit objects (or just a single one).

Once we migrate existing MRs to have this column, we will be able to get the
latest diff for a single merge request more efficiently, and (more importantly)
get all latest diffs for a collection of MRs efficiently.

MergeRequest#create_merge_request_diff and MergeRequest#reload_diff are
the only places where we generate a new MR diff so that's where we
should fetch the ref.

This also ensures that the ref is not fetched when we call
merge_request.merge_request_diffs.create in
Github::Import#fetch_pull_requests.

Signed-off-by: Rémy Coutable <remy@rymai.me>

Also, improved a bit the method names in
Github::Representation::PullRequest.

Last but not least, ensure temp branch names doesn't contain a `/` as
this would create the ref in a subfolder in `refs/heads` (e.g.
`refs/heads/gh-123/456/rymai/foo`), and would leave empty directories
upon branch deletion (e.g. `refs/heads/gh-123/456/rymai/`.

Signed-off-by: Rémy Coutable <remy@rymai.me>

update MR diff

Previously, we stored these as serialised fields - `st_{commits,diffs}` - on the
`merge_request_diffs` table. These now have their own tables -
`merge_request_diff_{commits,diffs}` - with a column for each attribute of the
serialised data.

Add a background migration to go through the existing MR diffs and migrate them
to the new format. Ignore any contents that cannot be displayed. Assuming that
we have 5 million rows to migrate, and each batch of 2,500 rows can be
completed in 5 minutes, this will take about 7 days to migrate everything.

Previously, we used Psych, which would:

1. Check if a string was encoded as binary, and not ASCII-compatible.
2. Add the !binary tag in that case.
3. Convert to base64.

We need to do the same thing, using a new column in place of the tag.