Add meaningful equality comparisons to many rust wrappers (#11401)

### What
pyo3 defaults `__eq__` to be equivalent to `is` but since we clone
before returning most python objects that means
`dataset.schema()==dataset.schema()` is False even though the contents
are identical.

I resolved this specific case and added a test.
Then I asked a robot to write me a lint to enforce this everywhere,
updated for things I could easily `derive(PartialEq, Eq)` then added
NOLINT to the rest.
The score is
| With eq | Without eq |
|---------|------------|
|      8       |      15           |

Author: ntjohnson1

rerun_py/src/catalog/catalog_client.rs

@@ -12,7 +12,7 @@ use crate::catalog::{
 };
 
 /// Client for a remote Rerun catalog server.
-#[pyclass(name = "CatalogClientInternal")]
+#[pyclass(name = "CatalogClientInternal")] // NOLINT: skip pyclass_eq, non-trivial implementation
 pub struct PyCatalogClientInternal {
     origin: re_uri::Origin,
 

rerun_py/src/catalog/dataframe_query.rs

@@ -23,7 +23,7 @@ use crate::catalog::{PyDatasetEntry, to_py_err};
 use crate::utils::{get_tokio_runtime, wait_for_future};
 
 /// View into a remote dataset acting as DataFusion table provider.
-#[pyclass(name = "DataframeQueryView")]
+#[pyclass(name = "DataframeQueryView")] // NOLINT: skip pyclass_eq, non-trivial implementation
 pub struct PyDataframeQueryView {
     dataset: Py<PyDatasetEntry>,
 

rerun_py/src/catalog/dataframe_rendering.rs

@@ -4,8 +4,8 @@ use pyo3::{Bound, PyAny, PyResult, pyclass, pymethods};
 
 use re_format_arrow::{RecordBatchFormatOpts, format_record_batch_opts};
 
-#[pyclass(name = "RerunHtmlTable")]
-#[derive(Clone)]
+#[pyclass(eq, name = "RerunHtmlTable")]
+#[derive(Clone, PartialEq, Eq)]
 pub struct PyRerunHtmlTable {
     max_width: Option<usize>,
     max_height: Option<usize>,

rerun_py/src/catalog/datafusion_table.rs

@@ -10,7 +10,7 @@ use tracing::instrument;
 use crate::catalog::PyCatalogClientInternal;
 use crate::utils::get_tokio_runtime;
 
-#[pyclass(frozen, name = "DataFusionTable")]
+#[pyclass(frozen, name = "DataFusionTable")] // NOLINT: skip pyclass_eq, non-trivial implementation
 pub struct PyDataFusionTable {
     pub provider: Arc<dyn TableProvider + Send>,
     pub name: String,

rerun_py/src/catalog/dataset_entry.rs

@@ -37,7 +37,7 @@ use super::{
 };
 
 /// A dataset entry in the catalog.
-#[pyclass(name = "DatasetEntry", extends=PyEntry)]
+#[pyclass(name = "DatasetEntry", extends=PyEntry)] // NOLINT: skip pyclass_eq, non-trivial implementation
 pub struct PyDatasetEntry {
     pub dataset_details: DatasetDetails,
     pub dataset_handle: DatasetHandle,

rerun_py/src/catalog/entry.rs

@@ -8,8 +8,8 @@ use re_protos::cloud::v1alpha1::{EntryKind, ext::EntryDetails};
 use crate::catalog::PyCatalogClientInternal;
 
 /// A unique identifier for an entry in the catalog.
-#[pyclass(name = "EntryId")]
-#[derive(Clone)]
+#[pyclass(eq, name = "EntryId")]
+#[derive(Clone, PartialEq, Eq)]
 pub struct PyEntryId {
     pub id: EntryId,
 }
@@ -104,7 +104,7 @@ impl From<PyEntryKind> for EntryKind {
 // ---
 
 /// An entry in the catalog.
-#[pyclass(name = "Entry", subclass)]
+#[pyclass(name = "Entry", subclass)] // NOLINT: skip pyclass_eq, non-trivial implementation
 pub struct PyEntry {
     pub client: Py<PyCatalogClientInternal>,
 

rerun_py/src/catalog/table_entry.rs

@@ -22,7 +22,7 @@ use crate::{
 ///
 /// Note: this object acts as a table provider for DataFusion.
 //TODO(ab): expose metadata about the table (e.g. stuff found in `provider_details`).
-#[pyclass(name = "TableEntry", extends=PyEntry)]
+#[pyclass(name = "TableEntry", extends=PyEntry)] // NOLINT: skip pyclass_eq, non-trivial implementation
 #[derive(Default)]
 pub struct PyTableEntry {
     lazy_provider: Option<Arc<dyn TableProvider + Send>>,

rerun_py/src/catalog/task.rs

@@ -11,7 +11,7 @@ use re_protos::common::v1alpha1::TaskId;
 use super::{PyCatalogClientInternal, PyDataFusionTable, to_py_err};
 
 /// A handle on a remote task.
-#[pyclass(name = "Task")]
+#[pyclass(name = "Task")] // NOLINT: skip pyclass_eq, non-trivial implementation
 pub struct PyTask {
     pub client: Py<PyCatalogClientInternal>,
 
@@ -48,7 +48,7 @@ impl PyTask {
 
 /// A collection of [`Task`].
 #[allow(rustdoc::broken_intra_doc_links)]
-#[pyclass(name = "Tasks")]
+#[pyclass(name = "Tasks")] // NOLINT: skip pyclass_eq, non-trivial implementation
 pub struct PyTasks {
     client: Py<PyCatalogClientInternal>,
 

rerun_py/src/dataframe/component_columns.rs

@@ -119,10 +119,11 @@ impl From<PyComponentColumnDescriptor> for ComponentColumnDescriptor {
 ///     The component to select
 #[pyclass(
     frozen,
+    eq,
     name = "ComponentColumnSelector",
     module = "rerun_bindings.rerun_bindings"
 )]
-#[derive(Clone)]
+#[derive(Clone, PartialEq, Eq)]
 pub struct PyComponentColumnSelector(pub ComponentColumnSelector);
 
 #[pymethods]

rerun_py/src/dataframe/index_columns.rs

@@ -11,10 +11,12 @@ use re_sorbet::{IndexColumnDescriptor, TimeColumnSelector};
 /// column, use [`IndexColumnSelector`][rerun.dataframe.IndexColumnSelector].
 #[pyclass(
     frozen,
+    eq,
+    hash,
     name = "IndexColumnDescriptor",
     module = "rerun_bindings.rerun_bindings"
 )]
-#[derive(Clone)]
+#[derive(Clone, PartialEq, Eq, Hash)]
 pub struct PyIndexColumnDescriptor(pub IndexColumnDescriptor);
 
 #[pymethods]
@@ -56,10 +58,11 @@ impl From<IndexColumnDescriptor> for PyIndexColumnDescriptor {
 ///     The name of the index to select. Usually the name of a timeline.
 #[pyclass(
     frozen,
+    eq,
     name = "IndexColumnSelector",
     module = "rerun_bindings.rerun_bindings"
 )]
-#[derive(Clone)]
+#[derive(Clone, PartialEq, Eq)]
 pub struct PyIndexColumnSelector(pub TimeColumnSelector);
 
 #[pymethods]