Create ruff_notebook crate (#7039)

## Summary

This PR moves `ruff/jupyter` into its own `ruff_notebook` crate. Beyond
the move itself, there were a few challenges:

1. `ruff_notebook` relies on the source map abstraction. I've moved the
source map into `ruff_diagnostics`, since it doesn't have any
dependencies on its own and is used alongside diagnostics.
2. `ruff_notebook` has a couple tests for end-to-end linting and
autofixing. I had to leave these tests in `ruff` itself.
3. We had code in `ruff/jupyter` that relied on Python lexing, in order
to provide a more targeted error message in the event that a user saves
a `.py` file with a `.ipynb` extension. I removed this in order to avoid
a dependency on the parser, it felt like it wasn't worth retaining just
for that dependency.

## Test Plan

`cargo test`

crates/ruff_notebook/src/schema.rs

@@ -0,0 +1,212 @@
+//! The JSON schema of a Jupyter Notebook, entrypoint is [`RawNotebook`]
+//!
+//! Generated by <https://app.quicktype.io/> from
+//! <https://github.com/jupyter/nbformat/blob/16b53251aabf472ad9406ddb1f78b0421c014eeb/nbformat/v4/nbformat.v4.schema.json>
+//! Jupyter Notebook v4.5 JSON schema.
+//!
+//! The following changes were made to the generated version:
+//! * Only keep the required structs and enums.
+//! * `Cell::id` is optional because it wasn't required <v4.5
+//! * `#[serde(deny_unknown_fields)]` was added where the schema had
+//!   `"additionalProperties": false`
+//! * `#[serde(flatten)] pub other: BTreeMap<String, Value>` for
+//!   `"additionalProperties": true` as preparation for round-trip support.
+//! * `#[serde(skip_serializing_none)]` was added to all structs where one or
+//!   more fields were optional to avoid serializing `null` values.
+//! * `Cell::execution_count` is a required property only for code cells, but
+//!   we serialize it for all cells. This is because we can't know if a cell is
+//!   a code cell or not without looking at the `cell_type` property, which
+//!   would require a custom serializer.
+
+use std::collections::BTreeMap;
+
+use serde::{Deserialize, Serialize};
+use serde_json::Value;
+use serde_with::skip_serializing_none;
+
+fn sort_alphabetically<T: Serialize, S: serde::Serializer>(
+    value: &T,
+    serializer: S,
+) -> Result<S::Ok, S::Error> {
+    let value = serde_json::to_value(value).map_err(serde::ser::Error::custom)?;
+    value.serialize(serializer)
+}
+
+/// This is used to serialize any value implementing [`Serialize`] alphabetically.
+///
+/// The reason for this is to maintain consistency in the generated JSON string,
+/// which is useful for diffing. The default serializer keeps the order of the
+/// fields as they are defined in the struct, which will not be consistent when
+/// there are `extra` fields.
+///
+/// # Example
+///
+/// ```
+/// use std::collections::BTreeMap;
+///
+/// use serde::Serialize;
+///
+/// use ruff_notebook::SortAlphabetically;
+///
+/// #[derive(Serialize)]
+/// struct MyStruct {
+///    a: String,
+///    #[serde(flatten)]
+///    extra: BTreeMap<String, String>,
+///    b: String,
+/// }
+///
+/// let my_struct = MyStruct {
+///     a: "a".to_string(),
+///     extra: BTreeMap::from([
+///         ("d".to_string(), "d".to_string()),
+///         ("c".to_string(), "c".to_string()),
+///     ]),
+///     b: "b".to_string(),
+/// };
+///
+/// let serialized = serde_json::to_string_pretty(&SortAlphabetically(&my_struct)).unwrap();
+/// assert_eq!(
+///     serialized,
+/// r#"{
+///   "a": "a",
+///   "b": "b",
+///   "c": "c",
+///   "d": "d"
+/// }"#
+/// );
+/// ```
+#[derive(Serialize)]
+pub struct SortAlphabetically<T: Serialize>(#[serde(serialize_with = "sort_alphabetically")] pub T);
+
+/// The root of the JSON of a Jupyter Notebook
+///
+/// Generated by <https://app.quicktype.io/> from
+/// <https://github.com/jupyter/nbformat/blob/16b53251aabf472ad9406ddb1f78b0421c014eeb/nbformat/v4/nbformat.v4.schema.json>
+/// Jupyter Notebook v4.5 JSON schema.
+#[derive(Clone, Debug, Serialize, Deserialize, PartialEq)]
+#[serde(deny_unknown_fields)]
+pub struct RawNotebook {
+    /// Array of cells of the current notebook.
+    pub cells: Vec<Cell>,
+    /// Notebook root-level metadata.
+    pub metadata: RawNotebookMetadata,
+    /// Notebook format (major number). Incremented between backwards incompatible changes to the
+    /// notebook format.
+    pub nbformat: i64,
+    /// Notebook format (minor number). Incremented for backward compatible changes to the
+    /// notebook format.
+    pub nbformat_minor: i64,
+}
+
+/// String identifying the type of cell.
+#[derive(Debug, Serialize, Deserialize, Clone, PartialEq)]
+#[serde(tag = "cell_type")]
+pub enum Cell {
+    #[serde(rename = "code")]
+    Code(CodeCell),
+    #[serde(rename = "markdown")]
+    Markdown(MarkdownCell),
+    #[serde(rename = "raw")]
+    Raw(RawCell),
+}
+
+/// Notebook raw nbconvert cell.
+#[skip_serializing_none]
+#[derive(Clone, Debug, Serialize, Deserialize, PartialEq)]
+#[serde(deny_unknown_fields)]
+pub struct RawCell {
+    pub attachments: Option<Value>,
+    /// Technically, id isn't required (it's not even present) in schema v4.0 through v4.4, but
+    /// it's required in v4.5. Main issue is that pycharm creates notebooks without an id
+    /// <https://youtrack.jetbrains.com/issue/PY-59438/Jupyter-notebooks-created-with-PyCharm-are-missing-the-id-field-in-cells-in-the-.ipynb-json>
+    pub id: Option<String>,
+    /// Cell-level metadata.
+    pub metadata: Value,
+    pub source: SourceValue,
+}
+
+/// Notebook markdown cell.
+#[skip_serializing_none]
+#[derive(Clone, Debug, Serialize, Deserialize, PartialEq)]
+#[serde(deny_unknown_fields)]
+pub struct MarkdownCell {
+    pub attachments: Option<Value>,
+    /// Technically, id isn't required (it's not even present) in schema v4.0 through v4.4, but
+    /// it's required in v4.5. Main issue is that pycharm creates notebooks without an id
+    /// <https://youtrack.jetbrains.com/issue/PY-59438/Jupyter-notebooks-created-with-PyCharm-are-missing-the-id-field-in-cells-in-the-.ipynb-json>
+    pub id: Option<String>,
+    /// Cell-level metadata.
+    pub metadata: Value,
+    pub source: SourceValue,
+}
+
+/// Notebook code cell.
+#[derive(Clone, Debug, Serialize, Deserialize, PartialEq)]
+#[serde(deny_unknown_fields)]
+pub struct CodeCell {
+    /// The code cell's prompt number. Will be null if the cell has not been run.
+    pub execution_count: Option<i64>,
+    /// Technically, id isn't required (it's not even present) in schema v4.0 through v4.4, but
+    /// it's required in v4.5. Main issue is that pycharm creates notebooks without an id
+    /// <https://youtrack.jetbrains.com/issue/PY-59438/Jupyter-notebooks-created-with-PyCharm-are-missing-the-id-field-in-cells-in-the-.ipynb-json>
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub id: Option<String>,
+    /// Cell-level metadata.
+    pub metadata: Value,
+    /// Execution, display, or stream outputs.
+    pub outputs: Vec<Value>,
+    pub source: SourceValue,
+}
+
+/// Notebook root-level metadata.
+#[skip_serializing_none]
+#[derive(Clone, Debug, Serialize, Deserialize, PartialEq)]
+pub struct RawNotebookMetadata {
+    /// The author(s) of the notebook document
+    pub authors: Option<Value>,
+    /// Kernel information.
+    pub kernelspec: Option<Value>,
+    /// Kernel information.
+    pub language_info: Option<LanguageInfo>,
+    /// Original notebook format (major number) before converting the notebook between versions.
+    /// This should never be written to a file.
+    pub orig_nbformat: Option<i64>,
+    /// The title of the notebook document
+    pub title: Option<String>,
+    /// For additional properties.
+    #[serde(flatten)]
+    pub extra: BTreeMap<String, Value>,
+}
+
+/// Kernel information.
+#[skip_serializing_none]
+#[derive(Clone, Debug, Serialize, Deserialize, PartialEq)]
+pub struct LanguageInfo {
+    /// The codemirror mode to use for code in this language.
+    pub codemirror_mode: Option<Value>,
+    /// The file extension for files in this language.
+    pub file_extension: Option<String>,
+    /// The mimetype corresponding to files in this language.
+    pub mimetype: Option<String>,
+    /// The programming language which this kernel runs.
+    pub name: String,
+    /// The pygments lexer to use for code in this language.
+    pub pygments_lexer: Option<String>,
+    /// For additional properties.
+    #[serde(flatten)]
+    pub extra: BTreeMap<String, Value>,
+}
+
+/// mimetype output (e.g. text/plain), represented as either an array of strings or a
+/// string.
+///
+/// Contents of the cell, represented as an array of lines.
+///
+/// The stream's text output, represented as an array of strings.
+#[derive(Clone, Debug, Serialize, Deserialize, PartialEq)]
+#[serde(untagged)]
+pub enum SourceValue {
+    String(String),
+    StringArray(Vec<String>),
+}