feat: linter for bare open (scoped) Classical

[grunweg]

Add a syntax linter for open Classical and open scoped Classical which are not scoped to a single declaration.
Such occurrences have been tracked as technical debt, and recently been fully removed from mathlib.
This linter enforces this (and removes the corresponding technical debt metric).

---

TODO: does the function for finding opens already exist elsewhere?

Second version of #15454.

[github-actions]

PR summary d0a98c7d58

Import changes for modified files

No significant changes to the import graph

Import changes for all files

Files	Import difference

---

Declarations diff

+ bar
+ baz
+ extractOpenNames
+ openClassicalLinter

You can run this locally as follows

## summary with just the declaration names:
./scripts/declarations_diff.sh <optional_commit>

## more verbose report:
./scripts/declarations_diff.sh long <optional_commit>

The doc-module for script/declarations_diff.sh contains some details about this script.

---

No changes to technical debt.

You can run this locally as

./scripts/technical-debt-metrics.sh pr_summary

• The relative value is the weighted sum of the differences with weight given by the inverse of the current value of the statistic.
• The absolute value is the relative value divided by the total sum of the inverses of the current values (i.e. the weighted average of the differences).

[adomani]

I remember this function already existing, but I cannot find it. Do you know if it already exists and if maybe it can be recycled?

[grunweg]

I don't remember seeing it, but have asked on zulip

[grunweg]

Status update: nobody responded on zulip --- perhaps it doesn't exist yet.

[jcommelin]

Please update the PR description at some point 😉

[grunweg]

@adomani Do you have an idea why the tests fail? They don't seem to register the new lint at all.

I'm honest at a loss as to why this occurs.

[adomani]

I think that there is a hardcoded test exception that should be MathlibTest instead.

[grunweg]

I think that there is a hardcoded test exception that should be MathlibTest instead.

🤦 Oh wow, thanks for pointing out the tomatoes on my eyes! Upon a closer look, I decided to remove those exclusions wholesale: downstream projects can decide to activate the linter in their lakefile.