[SymForce] Rename linearization_sparse_key_helper

Renaming `linearization_sparse_key_helper_t` to `linearization_offsets_t`.

I'd like to reuse this type for the dense linearizer, so I thought it
would be better to give it a name that is more suggestive of its
contents and is less specific about who is using it.

I'll be using it for more or less the exact same thing that the sparse
linearizer uses it for. (i.e., for figuring out where the components of
a factor's linearization get mapped to in the combined linearization).

Topic: rename_linearization_sparse_key_helper_t
Relative: fix_initial_residual_order
GitOrigin-RevId: adab6f10c6e3aaad66ca29de371db707a148a989

lcmtypes/symforce.lcm

@@ -93,7 +93,7 @@ struct linearization_dense_factor_helper_t {
   linearization_dense_key_helper_t key_helpers[];
 };
 
-struct linearization_sparse_key_helper_t {
+struct linearization_offsets_t {
   // Offset of this key within the factor's state vector
   int32_t factor_offset;
   // Offset of this key within the whole problem's state vector
@@ -108,7 +108,7 @@ struct linearization_sparse_factor_helper_t {
   // Offset of this factor's residual slice within the whole problem residual
   int32_t combined_residual_offset;
   // Data about each key's state offsets
-  linearization_sparse_key_helper_t key_helpers[];
+  linearization_offsets_t key_helpers[];
 
   // Mapping from factor jacobian flat storage into problem jacobian flat storage
   int32_t jacobian_index_map[];

symforce/opt/linearizer.cc

@@ -190,7 +190,7 @@ void Linearizer<ScalarType>::BuildInitialLinearization(const Values<Scalar>& val
 
       // Add contribution from right-hand-side
       for (int key_i = 0; key_i < static_cast<int>(factor_helper.key_helpers.size()); ++key_i) {
-        const linearization_sparse_key_helper_t& key_helper = factor_helper.key_helpers[key_i];
+        const linearization_offsets_t& key_helper = factor_helper.key_helpers[key_i];
 
         init_linearization_.rhs.segment(key_helper.combined_offset, key_helper.tangent_dim) +=
             linearized_factor.rhs.segment(key_helper.factor_offset, key_helper.tangent_dim);
@@ -385,7 +385,7 @@ void Linearizer<ScalarType>::UpdateFromLinearizedSparseFactorIntoSparse(
 
   // Add contribution from right-hand-side
   for (int key_i = 0; key_i < static_cast<int>(factor_helper.key_helpers.size()); ++key_i) {
-    const linearization_sparse_key_helper_t& key_helper = factor_helper.key_helpers[key_i];
+    const linearization_offsets_t& key_helper = factor_helper.key_helpers[key_i];
 
     linearization.rhs.segment(key_helper.combined_offset, key_helper.tangent_dim) +=
         linearized_factor.rhs.segment(key_helper.factor_offset, key_helper.tangent_dim);