Rename TensorSpecEncoding.kt class for Java callers (#400)

Adds `@file:JvmName("TensorSpecs")` to
`skainet-lang-core/.../tensor/ops/TensorSpecEncoding.kt`.

The file declares three top-level extension functions used to
read and write the `TensorEncoding` metadata that #469 plumbed
onto `TensorSpec`: `tensorEncoding`, `withTensorEncoding`, and
`inferTensorEncoding`. Top-level extensions in Kotlin compile to
static methods on a synthetic class named after the source file
— by default `TensorSpecEncodingKt`. Java call sites ended up
looking like:

    TensorEncoding encoding =
        TensorSpecEncodingKt.getTensorEncoding(spec);
    TensorSpec annotated =
        TensorSpecEncodingKt.withTensorEncoding(spec, TensorEncoding.Q8_0.INSTANCE);
    TensorEncoding data =
        TensorSpecEncodingKt.inferTensorEncoding(tensorData);

With `@file:JvmName("TensorSpecs")` they become:

    TensorEncoding encoding = TensorSpecs.getTensorEncoding(spec);
    TensorSpec annotated =
        TensorSpecs.withTensorEncoding(spec, TensorEncoding.Q8_0.INSTANCE);
    TensorEncoding data = TensorSpecs.inferTensorEncoding(tensorData);

Same Kotlin call sites are unaffected (they see the
top-level extension syntax either way) — `spec.tensorEncoding`
and `spec.withTensorEncoding(TensorEncoding.Q8_0)` still work
unchanged. Pure JVM-side binary name change.

Verified with jvmTest, compileKotlinWasmJs, macosArm64Test on
skainet-lang-core — no regressions.

Fourth of five commits polishing the Java / JVM consumption
story for the upcoming 0.19.0 release. See #400.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: michalharakal

skainet-lang/skainet-lang-core/src/commonMain/kotlin/sk/ainet/lang/tensor/ops/TensorSpecEncoding.kt

@@ -1,8 +1,11 @@
+@file:JvmName("TensorSpecs")
+
 package sk.ainet.lang.tensor.ops
 
 import sk.ainet.lang.tensor.data.TensorData
 import sk.ainet.lang.tensor.storage.PackedBlockStorage
 import sk.ainet.lang.tensor.storage.TensorEncoding
+import kotlin.jvm.JvmName
 
 /**
  * Metadata key used to carry a [TensorEncoding] on a [TensorSpec].